diff --git "a/trajectory/test.json" "b/trajectory/test.json" --- "a/trajectory/test.json" +++ "b/trajectory/test.json" @@ -1,3202 +1,3202 @@ [ - { - "id": "12329935", - "image": [ - "images/12329935_167.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(170, 47), (176, 77), (180, 84), (183, 90), (172, 84)]" - } - ] - }, - { - "id": 3516, - "image": [ - "images/3516_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"drag the cloth to the left side of the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(259, 274), (188, 296), (117, 299)]" - } - ] - }, - { - "id": "9899466", - "image": [ - "images/9899466_979.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(71, 92), (161, 32), (120, 45), (144, 36), (180, 43)]" - } - ] - }, - { - "id": "8287263", - "image": [ - "images/8287263_0290.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick wood cylinder. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(181, 21), (179, 22), (181, 24), (181, 22), (181, 22)]" - } - ] - }, - { - "id": 5933, - "image": [ - "images/5933_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the banana on the tray\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(103, 109), (195, 116), (273, 126), (362, 143)]" - } - ] - }, - { - "id": 1005, - "image": [ - "images/1005_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the blue spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(149, 156), (204, 132), (248, 166)]" - } - ] - }, - { - "id": 1712, - "image": [ - "images/1712_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the fork\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(282, 21), (273, 96), (273, 204), (280, 316)]" - } - ] - }, - { - "id": "3539090", - "image": [ - "images/3539090_0020.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(302, 48), (320, 126), (314, 230), (320, 123), (325, 45)]" - } - ] - }, - { - "id": 566, - "image": [ - "images/566_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the rag\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(327, 94), (254, 129), (183, 168), (112, 214)]" - } - ] - }, - { - "id": "33174307", - "image": [ - "images/33174307_15.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Stack the red cube on top of the green cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(128, 128), (122, 136), (124, 140), (133, 136), (143, 147)]" - } - ] - }, - { - "id": 3793, - "image": [ - "images/3793_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(113, 253), (103, 208), (99, 161), (107, 139)]" - } - ] - }, - { - "id": "25514701", - "image": [ - "images/25514701_108.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place the pepper in the ceramic bowl. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(66, 86), (67, 90), (63, 86), (92, 69), (105, 64)]" - } - ] - }, - { - "id": "11555052", - "image": [ - "images/11555052_564.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(180, 84), (181, 84), (181, 84), (178, 86), (176, 81)]" - } - ] - }, - { - "id": "17948694", - "image": [ - "images/17948694_184.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is open bottom drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(166, 207), (168, 211), (171, 207), (168, 219), (170, 215)]" - } - ] - }, - { - "id": "18332132", - "image": [ - "images/18332132_2.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is move orange near 7up can. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(288, 81), (197, 69), (192, 79), (128, 123), (221, 68)]" - } - ] - }, - { - "id": 4577, - "image": [ - "images/4577_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(120, 111), (161, 111), (198, 113), (242, 115), (281, 120), (335, 120)]" - } - ] - }, - { - "id": "27018718", - "image": [ - "images/27018718_137.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with sponge. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(80, 90), (73, 95), (69, 103), (70, 109), (67, 114)]" - } - ] - }, - { - "id": 3026, - "image": [ - "images/3026_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mushroom in the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(195, 99), (241, 105), (289, 118), (341, 164)]" - } - ] - }, - { - "id": "18237860", - "image": [ - "images/18237860_16.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is pick blue plastic bottle. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(85, 119), (89, 119), (80, 108), (67, 89), (64, 82)]" - } - ] - }, - { - "id": "22723644", - "image": [ - "images/22723644_2.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(472, 302), (392, 330), (376, 342), (284, 290), (222, 308)]" - } - ] - }, - { - "id": 4268, - "image": [ - "images/4268_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the button\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(118, 209), (113, 186), (112, 161)]" - } - ] - }, - { - "id": "4325508", - "image": [ - "images/4325508_0162.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(62, 196), (55, 170), (224, 168), (437, 264), (478, 44)]" - } - ] - }, - { - "id": "15858295", - "image": [ - "images/15858295_10.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is move the blue triangle towards the yellow hexagon. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(195, 147), (179, 117), (273, 146), (362, 120)]" - } - ] - }, - { - "id": 2, - "image": [ - "images/2_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the croissant\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(104, 33), (97, 39), (92, 50), (88, 59)]" - } - ] - }, - { - "id": 6037, - "image": [ - "images/6037_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(131, 114), (181, 120), (223, 136), (270, 154)]" - } - ] - }, - { - "id": "11372736", - "image": [ - "images/11372736_230.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(178, 79), (180, 81), (181, 86), (181, 79), (176, 81)]" - } - ] - }, - { - "id": "11720865", - "image": [ - "images/11720865_406.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(187, 101), (191, 109), (180, 101), (196, 109), (200, 96)]" - } - ] - }, - { - "id": "10443781", - "image": [ - "images/10443781_149.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(183, 99), (180, 99), (183, 99), (189, 99), (180, 99)]" - } - ] - }, - { - "id": 5129, - "image": [ - "images/5129_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the strainer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(103, 116), (117, 103), (131, 102), (159, 102)]" - } - ] - }, - { - "id": 5970, - "image": [ - "images/5970_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(119, 111), (164, 118), (203, 125), (242, 139)]" - } - ] - }, - { - "id": 2954, - "image": [ - "images/2954_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(248, 96), (218, 237), (241, 277)]" - } - ] - }, - { - "id": "13473464", - "image": [ - "images/13473464_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a WidowX robot using the end effector control. The task is take carrot off plate cardboardfence. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(209, 172), (186, 244), (156, 280), (224, 64), (170, 98)]" - } - ] - }, - { - "id": "9103065", - "image": [ - "images/9103065_0385.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick red green blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(300, 149), (302, 144), (302, 144), (300, 140), (300, 140)]" - } - ] - }, - { - "id": 6143, - "image": [ - "images/6143_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(125, 118), (180, 109), (228, 94), (270, 70)]" - } - ] - }, - { - "id": "17059094", - "image": [ - "images/17059094_15.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is slide green circle on top of red circle and hexagon. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(341, 51), (328, 36), (247, 41), (282, 39), (284, 33)]" - } - ] - }, - { - "id": 6192, - "image": [ - "images/6192_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(111, 112), (150, 90), (184, 82), (222, 79)]" - } - ] - }, - { - "id": 1836, - "image": [ - "images/1836_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(252, 170), (288, 187), (304, 211), (304, 247)]" - } - ] - }, - { - "id": 6198, - "image": [ - "images/6198_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the banana on the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(116, 114), (167, 130), (208, 132), (261, 130)]" - } - ] - }, - { - "id": "11405107", - "image": [ - "images/11405107_51.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(131, 127), (176, 68), (168, 109), (172, 122), (196, 114)]" - } - ] - }, - { - "id": 5719, - "image": [ - "images/5719_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(156, 116), (169, 128), (175, 136), (178, 150)]" - } - ] - }, - { - "id": "14718878", - "image": [ - "images/14718878_24.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is place the yellow heart horizontally left to the red circle. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(188, 99), (218, 105), (197, 105), (256, 77), (259, 65)]" - } - ] - }, - { - "id": 2271, - "image": [ - "images/2271_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"lower the gripper towards the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(97, 213), (174, 222), (248, 228), (328, 250), (398, 302)]" - } - ] - }, - { - "id": 6046, - "image": [ - "images/6046_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle in the strainer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(131, 108), (139, 126), (153, 144), (170, 155)]" - } - ] - }, - { - "id": 809, - "image": [ - "images/809_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the pan\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(231, 117), (312, 126), (373, 134), (488, 223)]" - } - ] - }, - { - "id": "28268704", - "image": [ - "images/28268704_90.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place the eraser on the white sponge. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(103, 70), (105, 70), (103, 70), (105, 68), (103, 69)]" - } - ] - }, - { - "id": 3549, - "image": [ - "images/3549_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the object in the first container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(371, 223), (289, 244), (261, 259)]" - } - ] - }, - { - "id": "21487705", - "image": [ - "images/21487705_5.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(437, 356), (433, 306), (432, 332), (440, 335), (460, 240)]" - } - ] - }, - { - "id": "21948875", - "image": [ - "images/21948875_15.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(252, 257), (250, 251), (241, 306), (273, 172)]" - } - ] - }, - { - "id": 4613, - "image": [ - "images/4613_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(142, 109), (164, 121), (198, 128), (239, 132), (262, 136)]" - } - ] - }, - { - "id": 971, - "image": [ - "images/971_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the can\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(277, 226), (382, 245), (432, 212)]" - } - ] - }, - { - "id": "15047872", - "image": [ - "images/15047872_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is move the blue triangle towards the top left. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(323, 23), (254, 23), (231, 18), (195, 18), (195, 18)]" - } - ] - }, - { - "id": 2078, - "image": [ - "images/2078_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(64, 226), (161, 204), (227, 184), (252, 176)]" - } - ] - }, - { - "id": 6823, - "image": [ - "images/6823_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(116, 109), (158, 109), (208, 115), (245, 128)]" - } - ] - }, - { - "id": 6666, - "image": [ - "images/6666_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(131, 119), (145, 132), (156, 139), (170, 149)]" - } - ] - }, - { - "id": "12242369", - "image": [ - "images/12242369_822.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(172, 75), (69, 66), (165, 34), (170, 49), (165, 45)]" - } - ] - }, - { - "id": "17456351", - "image": [ - "images/17456351_46.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is move paper bowl near apple. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(160, 11), (161, 11), (169, 17), (166, 35), (163, 33)]" - } - ] - }, - { - "id": "26219537", - "image": [ - "images/26219537_136.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with eraser. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(79, 120), (79, 120), (80, 120), (79, 120), (79, 121)]" - } - ] - }, - { - "id": "18829288", - "image": [ - "images/18829288_48.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is open top drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(206, 152), (193, 150), (193, 167), (211, 179), (230, 178)]" - } - ] - }, - { - "id": 4026, - "image": [ - "images/4026_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the button\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(115, 206), (120, 133), (125, 125)]" - } - ] - }, - { - "id": 1687, - "image": [ - "images/1687_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the potato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(293, 139), (309, 219), (316, 321), (320, 410)]" - } - ] - }, - { - "id": "24148424", - "image": [ - "images/24148424_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(353, 334), (352, 353), (375, 292), (376, 302), (353, 366)]" - } - ] - }, - { - "id": "14189282", - "image": [ - "images/14189282_16.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is push the blue triangle towards the yellow heart. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(476, 42), (467, 38), (465, 38), (366, 45)]" - } - ] - }, - { - "id": "25596583", - "image": [ - "images/25596583_130.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place pepper in metal cup. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(70, 99), (61, 97), (59, 101), (69, 87), (70, 86)]" - } - ] - }, - { - "id": 364, - "image": [ - "images/364_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"pick up the green cup\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(189, 8), (168, 12), (129, 17), (105, 21), (86, 28), (45, 40), (30, 42)]" - } - ] - }, - { - "id": "27685269", - "image": [ - "images/27685269_86.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place towel in table surface. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(94, 75), (94, 75), (94, 75), (94, 74), (94, 75)]" - } - ] - }, - { - "id": "3505702", - "image": [ - "images/3505702_0408.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(414, 132), (410, 219), (410, 232), (432, 60), (440, 34)]" - } - ] - }, - { - "id": 6298, - "image": [ - "images/6298_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(113, 112), (169, 138), (203, 148), (233, 130)]" - } - ] - }, - { - "id": "796755", - "image": [ - "images/796755_1465.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Interact with the objects in diverse but meaningful ways. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(40, 44), (70, 8), (24, 24), (70, 9), (72, 13)]" - } - ] - }, - { - "id": 599, - "image": [ - "images/599_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(85, 189), (163, 211), (240, 241), (337, 309)]" - } - ] - }, - { - "id": 5234, - "image": [ - "images/5234_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"pick up the lid of the tea container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(214, 4), (250, 17), (292, 43), (312, 66)]" - } - ] - }, - { - "id": 4016, - "image": [ - "images/4016_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the cabinet door handles\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(120, 208), (144, 113), (136, 102)]" - } - ] - }, - { - "id": 1325, - "image": [ - "images/1325_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the stuffed animal\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(350, 15), (394, 107), (448, 205), (467, 310)]" - } - ] - }, - { - "id": "3912066", - "image": [ - "images/3912066_0006.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is destack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(336, 43), (300, 216), (323, 181), (396, 373), (416, 116)]" - } - ] - }, - { - "id": "20373004", - "image": [ - "images/20373004_3.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is open middle drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(281, 115), (240, 214), (238, 214), (254, 221), (262, 216)]" - } - ] - }, - { - "id": 5045, - "image": [ - "images/5045_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(106, 106), (161, 125), (226, 130), (292, 120)]" - } - ] - }, - { - "id": 5343, - "image": [ - "images/5343_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(137, 115), (215, 134), (256, 134)]" - } - ] - }, - { - "id": "12495911", - "image": [ - "images/12495911_185.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble lamp. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(163, 42), (172, 45), (170, 47), (71, 36), (159, 19)]" - } - ] - }, - { - "id": "14160160", - "image": [ - "images/14160160_11.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is push the red circle and green circle to the left side of the board. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(296, 24), (241, 30), (245, 24), (261, 20), (256, 28)]" - } - ] - }, - { - "id": 4532, - "image": [ - "images/4532_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(136, 108), (175, 121), (220, 128), (272, 132), (293, 136)]" - } - ] - }, - { - "id": "1759160", - "image": [ - "images/1759160_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(133, 25), (208, 36), (178, 63), (189, 60), (170, 70)]" - } - ] - }, - { - "id": "5601041", - "image": [ - "images/5601041_0114.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(444, 103), (432, 175), (426, 194), (449, 82), (451, 43)]" - } - ] - }, - { - "id": "12710602", - "image": [ - "images/12710602_85.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(178, 90), (168, 68), (187, 103), (174, 116), (198, 140)]" - } - ] - }, - { - "id": "20396683", - "image": [ - "images/20396683_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is open middle drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(281, 187), (224, 192), (224, 189), (226, 211), (240, 199)]" - } - ] - }, - { - "id": "30235189", - "image": [ - "images/30235189_231.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Turn on the faucet by rotating a designated handle. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(182, 56), (184, 56), (186, 57), (188, 55)]" - } - ] - }, - { - "id": 6528, - "image": [ - "images/6528_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"pull the strainer backwards\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(159, 131), (155, 150), (156, 178), (164, 194)]" - } - ] - }, - { - "id": "24626661", - "image": [ - "images/24626661_2.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(407, 291), (421, 300), (396, 328), (423, 300), (476, 243)]" - } - ] - }, - { - "id": 781, - "image": [ - "images/781_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the bowl\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(129, 103), (188, 134), (227, 175), (245, 252)]" - } - ] - }, - { - "id": 5496, - "image": [ - "images/5496_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move the mug to the right\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(198, 148), (222, 149), (251, 144), (268, 138)]" - } - ] - }, - { - "id": 2561, - "image": [ - "images/2561_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move the square above the fork\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(177, 208), (211, 187), (268, 180), (311, 187)]" - } - ] - }, - { - "id": "32148539", - "image": [ - "images/32148539_73.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(109, 122), (110, 128), (109, 129), (112, 118), (114, 109)]" - } - ] - }, - { - "id": "25926531", - "image": [ - "images/25926531_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is drag the pepper across the table. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(135, 62), (122, 63), (102, 82), (90, 97), (48, 99)]" - } - ] - }, - { - "id": 587, - "image": [ - "images/587_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(444, 197), (360, 177), (282, 160), (243, 163)]" - } - ] - }, - { - "id": "17651079", - "image": [ - "images/17651079_31.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is open top drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(148, 218), (148, 224), (149, 228), (154, 231)]" - } - ] - }, - { - "id": "7842628", - "image": [ - "images/7842628_0091.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack green blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(295, 173), (270, 348), (323, 211), (380, 216), (433, 26)]" - } - ] - }, - { - "id": "9980701", - "image": [ - "images/9980701_119.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(180, 68), (105, 86), (176, 69), (176, 62), (202, 68)]" - } - ] - }, - { - "id": "12998999", - "image": [ - "images/12998999_45.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Insert the blue gear onto the right peg, followed by the red gear. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(63, 13), (59, 15), (47, 24), (47, 6), (44, 26)]" - } - ] - }, - { - "id": "24256186", - "image": [ - "images/24256186_5.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(483, 352), (485, 352), (483, 351), (476, 329), (467, 348)]" - } - ] - }, - { - "id": 2158, - "image": [ - "images/2158_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"place the banana next to the bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(177, 99), (213, 173), (238, 248), (272, 311)]" - } - ] - }, - { - "id": 5142, - "image": [ - "images/5142_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(133, 114), (144, 134), (164, 147), (192, 149)]" - } - ] - }, - { - "id": "15979875", - "image": [ - "images/15979875_11.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is push yellow hexagon along with the green circle to the right side of the board. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(240, 128), (305, 137), (343, 137), (419, 136)]" - } - ] - }, - { - "id": "12017587", - "image": [ - "images/12017587_481.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(191, 71), (189, 71), (187, 75), (187, 73), (185, 73)]" - } - ] - }, - { - "id": 323, - "image": [ - "images/323_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"lift the sponge from the blue container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(211, 77), (195, 60), (170, 56), (133, 68)]" - } - ] - }, - { - "id": "24169703", - "image": [ - "images/24169703_6.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(327, 309), (330, 290), (364, 249), (348, 290), (311, 306)]" - } - ] - }, - { - "id": "730810", - "image": [ - "images/730810_885.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Interact with the objects in diverse but meaningful ways. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(71, 16), (107, 34), (64, 19), (29, 86), (27, 93)]" - } - ] - }, - { - "id": "32462209", - "image": [ - "images/32462209_105.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Insert the peg into the horizontal hole in a box. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(124, 96), (125, 93), (128, 96), (130, 91), (130, 90)]" - } - ] - }, - { - "id": "17584409", - "image": [ - "images/17584409_23.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is place 7up can upright. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(196, 64), (200, 67), (201, 69), (214, 81), (232, 88)]" - } - ] - }, - { - "id": "6252498", - "image": [ - "images/6252498_0205.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack gray white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(272, 80), (273, 156), (275, 144), (273, 44), (277, 29)]" - } - ] - }, - { - "id": "28418885", - "image": [ - "images/28418885_74.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is wipe red bowl with eraser. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(165, 100), (116, 80), (119, 81), (119, 81), (130, 86)]" - } - ] - }, - { - "id": "2023203", - "image": [ - "images/2023203_21.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(201, 37), (186, 37), (179, 36), (177, 42), (181, 33)]" - } - ] - }, - { - "id": 4793, - "image": [ - "images/4793_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(114, 132), (195, 102), (242, 81), (272, 61)]" - } - ] - }, - { - "id": "1959009", - "image": [ - "images/1959009_19.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(96, 109), (81, 116), (129, 111), (173, 144), (176, 121)]" - } - ] - }, - { - "id": "12213000", - "image": [ - "images/12213000_1141.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble square_table. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(133, 116), (170, 79), (180, 75), (112, 75), (103, 71)]" - } - ] - }, - { - "id": 3112, - "image": [ - "images/3112_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move the corn towards the middle right of the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(590, 79), (554, 97), (519, 132), (494, 176), (485, 213), (485, 261), (485, 294)]" - } - ] - }, - { - "id": "28220565", - "image": [ - "images/28220565_79.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place pepper in purple bowl. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(145, 91), (145, 90), (135, 74), (122, 70), (122, 71)]" - } - ] - }, - { - "id": "16581704", - "image": [ - "images/16581704_24.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is move the yellow hexagon diagonally right to the green circle. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(378, 212), (371, 208), (378, 208), (380, 189), (405, 182)]" - } - ] - }, - { - "id": "31019165", - "image": [ - "images/31019165_60.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(109, 154), (108, 154), (115, 150), (168, 119), (192, 103)]" - } - ] - }, - { - "id": 2131, - "image": [ - "images/2131_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the sushi roll\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(321, 17), (302, 84), (282, 140), (266, 204)]" - } - ] - }, - { - "id": "25145241", - "image": [ - "images/25145241_90.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place eraser in tray. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(119, 82), (116, 58), (112, 53), (118, 66), (119, 64)]" - } - ] - }, - { - "id": "30600931", - "image": [ - "images/30600931_18.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(123, 124), (122, 143), (122, 154), (121, 132), (123, 43)]" - } - ] - }, - { - "id": "5836113", - "image": [ - "images/5836113_0211.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack boston blue white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(419, 340), (403, 240), (289, 164), (188, 144), (176, 24)]" - } - ] - }, - { - "id": "5759626", - "image": [ - "images/5759626_0256.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack purple white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(97, 202), (163, 153), (307, 142), (336, 208), (343, 34)]" - } - ] - }, - { - "id": 2363, - "image": [ - "images/2363_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the banana in the colander\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(291, 103), (328, 136), (352, 189), (360, 245), (373, 342)]" - } - ] - }, - { - "id": "28833023", - "image": [ - "images/28833023_36.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place sponge in tray. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(69, 77), (50, 95), (48, 101), (51, 104), (51, 107)]" - } - ] - }, - { - "id": "8810231", - "image": [ - "images/8810231_0157.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick red green cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(392, 245), (394, 210), (396, 134), (405, 54), (407, 46)]" - } - ] - }, - { - "id": 430, - "image": [ - "images/430_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move the bread towards the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(77, 42), (64, 38), (53, 40), (40, 45)]" - } - ] - }, - { - "id": 3365, - "image": [ - "images/3365_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the yellow brush\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(248, 142), (273, 184), (302, 236), (293, 286)]" - } - ] - }, - { - "id": 4231, - "image": [ - "images/4231_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the cabinet door\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(123, 204), (123, 184), (112, 164), (101, 152), (79, 132)]" - } - ] - }, - { - "id": 138, - "image": [ - "images/138_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(36, 36), (43, 49), (51, 58), (66, 68)]" - } - ] - }, - { - "id": "7288602", - "image": [ - "images/7288602_0020.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick red green blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(293, 39), (227, 85), (216, 162), (220, 190), (211, 115)]" - } - ] - }, - { - "id": "2479394", - "image": [ - "images/2479394_25.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(171, 64), (182, 58), (190, 58), (197, 48), (198, 50)]" - } - ] - }, - { - "id": "5930764", - "image": [ - "images/5930764_0174.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(316, 197), (320, 168), (321, 115), (325, 85), (328, 38)]" - } - ] - }, - { - "id": 4275, - "image": [ - "images/4275_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the drawer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(110, 198), (119, 163), (125, 130), (135, 87)]" - } - ] - }, - { - "id": "22157960", - "image": [ - "images/22157960_11.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(405, 311), (364, 329), (353, 352), (357, 339), (437, 244)]" - } - ] - }, - { - "id": 5894, - "image": [ - "images/5894_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(113, 108), (128, 131), (145, 150), (170, 165)]" - } - ] - }, - { - "id": "28570949", - "image": [ - "images/28570949_29.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is wipe table surface with towel. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(172, 104), (142, 122), (133, 121), (119, 110), (122, 112)]" - } - ] - }, - { - "id": 639, - "image": [ - "images/639_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(96, 49), (138, 81), (184, 121), (229, 155)]" - } - ] - }, - { - "id": 4707, - "image": [ - "images/4707_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move the mug to the right\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(212, 147), (231, 139), (250, 144), (262, 157)]" - } - ] - }, - { - "id": 4197, - "image": [ - "images/4197_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the microwave oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(107, 215), (141, 165), (174, 98)]" - } - ] - }, - { - "id": "6803675", - "image": [ - "images/6803675_0229.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is destack boston blue gray cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(136, 111), (197, 157), (296, 257), (323, 310), (323, 106)]" - } - ] - }, - { - "id": 2764, - "image": [ - "images/2764_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"place the apple on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(179, 217), (188, 151), (186, 96), (177, 32)]" - } - ] - }, - { - "id": 1323, - "image": [ - "images/1323_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the spatula\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(142, 56), (193, 78), (248, 142), (272, 252)]" - } - ] - }, - { - "id": "31323233", - "image": [ - "images/31323233_61.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Insert a designated object into the corresponding slot on a board. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(162, 88), (152, 90), (93, 105), (52, 120), (55, 122)]" - } - ] - }, - { - "id": "13778353", - "image": [ - "images/13778353_0.jpg" - ], - "conversations": [ - { - "value": "\n You are a WidowX robot using the end effector control. The task is move towel to lower right corner. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(256, 110), (362, 147), (392, 173), (435, 303), (451, 245)]" - } - ] - }, - { - "id": "1512170", - "image": [ - "images/1512170_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(105, 94), (114, 96), (104, 66), (96, 63), (141, 61)]" - } - ] - }, - { - "id": 1153, - "image": [ - "images/1153_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the tomato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(277, 105), (279, 128), (270, 152), (245, 96)]" - } - ] - }, - { - "id": "12633235", - "image": [ - "images/12633235_252.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(12, 152), (12, 152), (12, 152), (185, 105), (12, 152)]" - } - ] - }, - { - "id": "15944740", - "image": [ - "images/15944740_8.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is separate the blue blocks. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(49, 87), (128, 182), (243, 267), (464, 244), (483, 227)]" - } - ] - }, - { - "id": 4746, - "image": [ - "images/4746_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(144, 119), (178, 116), (220, 116), (251, 120)]" - } - ] - }, - { - "id": "8171470", - "image": [ - "images/8171470_0129.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick 4wp wood cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(229, 170), (229, 202), (231, 204), (232, 156), (225, 49)]" - } - ] - }, - { - "id": "8022683", - "image": [ - "images/8022683_0084.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack green blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(250, 139), (213, 254), (272, 124), (296, 181), (300, 15)]" - } - ] - }, - { - "id": 2512, - "image": [ - "images/2512_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(259, 48), (199, 85), (133, 146), (85, 216)]" - } - ] - }, - { - "id": "23048", - "image": [ - "images/23048_74.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Kindly get your arm to the topmost drawer handle.\nGet a good grip on it, then softly pull to open it. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(86, 34), (102, 29), (104, 32), (96, 35), (91, 39)]" - } - ] - }, - { - "id": "10559325", - "image": [ - "images/10559325_1694.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(114, 34), (107, 34), (112, 32), (96, 36), (98, 36)]" - } - ] - }, - { - "id": 4787, - "image": [ - "images/4787_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(130, 114), (172, 116), (206, 119), (257, 130)]" - } - ] - }, - { - "id": 357, - "image": [ - "images/357_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"pick up the breadstick\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(146, 122), (133, 99), (111, 81), (88, 71), (66, 62), (45, 60)]" - } - ] - }, - { - "id": "3464533", - "image": [ - "images/3464533_0047.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(234, 84), (103, 169), (83, 240), (71, 192), (44, 84)]" - } - ] - }, - { - "id": "29094723", - "image": [ - "images/29094723_72.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is place towel in plastic cup. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(57, 75), (51, 68), (56, 71), (56, 73), (57, 70)]" - } - ] - }, - { - "id": 3959, - "image": [ - "images/3959_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"approach the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(130, 227), (133, 194), (132, 173), (130, 140), (124, 115), (115, 90)]" - } - ] - }, - { - "id": "22146877", - "image": [ - "images/22146877_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(476, 225), (481, 244), (492, 242), (471, 235), (401, 269)]" - } - ] - }, - { - "id": "6521133", - "image": [ - "images/6521133_0242.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is stack gray white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(204, 105), (206, 144), (204, 108), (195, 41), (241, 12)]" - } - ] - }, - { - "id": 5155, - "image": [ - "images/5155_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(156, 160), (178, 144), (206, 136), (236, 130)]" - } - ] - }, - { - "id": 1124, - "image": [ - "images/1124_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the cabinet handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(528, 256), (512, 297), (476, 328), (439, 347)]" - } - ] - }, - { - "id": 4445, - "image": [ - "images/4445_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"hover over the round object\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(123, 3), (122, 15), (119, 33), (119, 48)]" - } - ] - }, - { - "id": "14248106", - "image": [ - "images/14248106_11.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is push the red circle to the top of the yellow hexagon. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(74, 55), (368, 50), (449, 39), (464, 55), (412, 68)]" - } - ] - }, - { - "id": "20796777", - "image": [ - "images/20796777_13.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(563, 239), (545, 307), (533, 227), (524, 305), (540, 309)]" - } - ] - }, - { - "id": "15124148", - "image": [ - "images/15124148_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is move the green star into the green circle. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(129, 50), (154, 50), (163, 48), (167, 51), (145, 47)]" - } - ] - }, - { - "id": 6186, - "image": [ - "images/6186_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(114, 114), (192, 107), (262, 114), (348, 134)]" - } - ] - }, - { - "id": 6444, - "image": [ - "images/6444_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the bottle on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(155, 116), (203, 105), (231, 78)]" - } - ] - }, - { - "id": 528, - "image": [ - "images/528_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the bowl\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(426, 158), (385, 189), (359, 213), (328, 241)]" - } - ] - }, - { - "id": "21124308", - "image": [ - "images/21124308_11.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(456, 332), (417, 365), (432, 352), (400, 392), (405, 357)]" - } - ] - }, - { - "id": "12641200", - "image": [ - "images/12641200_545.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble desk. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(73, 51), (118, 75), (176, 62), (73, 58), (77, 71)]" - } - ] - }, - { - "id": 6205, - "image": [ - "images/6205_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(117, 118), (176, 120), (228, 132), (267, 152)]" - } - ] - }, - { - "id": "10348867", - "image": [ - "images/10348867_199.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is assemble stool. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(172, 64), (79, 101), (152, 51), (150, 53), (142, 64)]" - } - ] - }, - { - "id": "21975924", - "image": [ - "images/21975924_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(506, 311), (506, 314), (508, 314), (506, 314), (508, 312)]" - } - ] - }, - { - "id": "8764635", - "image": [ - "images/8764635_0255.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is destack red blue cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(385, 340), (378, 360), (385, 319), (389, 206), (400, 156)]" - } - ] - }, - { - "id": 6337, - "image": [ - "images/6337_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(128, 112), (175, 138), (223, 143), (265, 138)]" - } - ] - }, - { - "id": "1555635", - "image": [ - "images/1555635_13.jpg" - ], - "conversations": [ - { - "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(230, 70), (210, 68), (194, 64), (225, 54), (214, 84)]" - } - ] - }, - { - "id": 675, - "image": [ - "images/675_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"grip the vessel\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(213, 76), (257, 164), (291, 228), (328, 276)]" - } - ] - }, - { - "id": "17598097", - "image": [ - "images/17598097_23.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is pick orange from white bowl. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(165, 65), (189, 50), (214, 50), (216, 79)]" - } - ] - }, - { - "id": "28501760", - "image": [ - "images/28501760_2.jpg" - ], - "conversations": [ - { - "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with sponge. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(63, 73), (63, 74), (60, 69), (44, 69), (51, 70)]" - } - ] - }, - { - "id": 5111, - "image": [ - "images/5111_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(142, 114), (173, 121), (203, 127), (251, 139)]" - } - ] - }, - { - "id": "20068889", - "image": [ - "images/20068889_12.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is move green can near redbull can. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(144, 45), (122, 45), (161, 78), (115, 120), (234, 93)]" - } - ] - }, - { - "id": 4819, - "image": [ - "images/4819_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"drag the mug forward\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(262, 114), (270, 83), (259, 58), (262, 33)]" - } - ] - }, - { - "id": 1927, - "image": [ - "images/1927_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the potato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(55, 84), (83, 117), (113, 130), (172, 128)]" - } - ] - }, - { - "id": 4281, - "image": [ - "images/4281_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(108, 215), (119, 173), (119, 128), (119, 105)]" - } - ] - }, - { - "id": "18398623", - "image": [ - "images/18398623_1.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is pick orange can from middle drawer and place on counter. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(296, 88), (234, 187), (192, 189), (216, 48), (222, 58)]" - } - ] - }, - { - "id": "7602610", - "image": [ - "images/7602610_0221.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is destack gray white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(170, 204), (170, 202), (165, 156), (154, 73), (152, 48)]" - } - ] - }, - { - "id": 6012, - "image": [ - "images/6012_frame_15.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"close the drawer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(181, 93), (170, 102), (161, 118), (151, 130)]" - } - ] - }, - { - "id": 143, - "image": [ - "images/143_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"move towards the red object\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(43, 96), (68, 103), (83, 111), (96, 118), (99, 133)]" - } - ] - }, - { - "id": "5440609", - "image": [ - "images/5440609_0241.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the joint control. The task is pick gray white cube. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(353, 45), (353, 46), (352, 46), (353, 46), (352, 46)]" - } - ] - }, - { - "id": "15039374", - "image": [ - "images/15039374_10.jpg" - ], - "conversations": [ - { - "value": "\n You are a xArm robot using the end effector control. The task is nudge the red circle right a bit. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(298, 20), (296, 30), (323, 20), (346, 29)]" - } - ] - }, - { - "id": "33190562", - "image": [ - "images/33190562_9.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(122, 121), (121, 124), (122, 135), (122, 135), (98, 61)]" - } - ] - }, - { - "id": "21153184", - "image": [ - "images/21153184_13.jpg" - ], - "conversations": [ - { - "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(508, 280), (503, 303), (499, 311), (533, 345), (483, 390)]" - } - ] - }, - { - "id": "19262395", - "image": [ - "images/19262395_7.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is close middle drawer. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(224, 224), (216, 226), (201, 235), (202, 194), (208, 186)]" - } - ] - }, - { - "id": 5012, - "image": [ - "images/5012_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(150, 124), (228, 121), (251, 138)]" - } - ] - }, - { - "id": 3224, - "image": [ - "images/3224_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(147, 66), (206, 118), (270, 180), (293, 267)]" - } - ] - }, - { - "id": 5928, - "image": [ - "images/5928_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(109, 112), (153, 121), (200, 130), (251, 132)]" - } - ] - }, - { - "id": "32557112", - "image": [ - "images/32557112_110.jpg" - ], - "conversations": [ - { - "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", - "from": "human" - }, - { - "from": "gpt", - "value": "[(115, 104), (108, 84), (96, 60), (88, 38), (80, 37)]" - } - ] - }, - { - "id": 3868, - "image": [ - "images/3868_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"reach for the bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(123, 174), (99, 115), (116, 81)]" - } - ] - }, - { - "id": 543, - "image": [ - "images/543_frame_0.png" - ], - "conversations": [ - { - "from": "human", - "value": "\nYou are a robot using the joint control. The task is \"pick up the pan\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." - }, - { - "from": "gpt", - "value": "[(206, 48), (211, 127), (225, 215), (248, 292)]" - } - ] - } + { + "id": 12329935, + "image": [ + "images/12329935_167.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(170, 47), (176, 77), (180, 84), (183, 90), (172, 84)]" + } + ] + }, + { + "id": 3516, + "image": [ + "images/3516_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"drag the cloth to the left side of the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(259, 274), (188, 296), (117, 299)]" + } + ] + }, + { + "id": 9899466, + "image": [ + "images/9899466_979.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(71, 92), (161, 32), (120, 45), (144, 36), (180, 43)]" + } + ] + }, + { + "id": 8287263, + "image": [ + "images/8287263_0290.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick wood cylinder. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(181, 21), (179, 22), (181, 24), (181, 22), (181, 22)]" + } + ] + }, + { + "id": 5933, + "image": [ + "images/5933_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the banana on the tray\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(103, 109), (195, 116), (273, 126), (362, 143)]" + } + ] + }, + { + "id": 1005, + "image": [ + "images/1005_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the blue spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(149, 156), (204, 132), (248, 166)]" + } + ] + }, + { + "id": 1712, + "image": [ + "images/1712_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the fork\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(282, 21), (273, 96), (273, 204), (280, 316)]" + } + ] + }, + { + "id": 3539090, + "image": [ + "images/3539090_0020.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(302, 48), (320, 126), (314, 230), (320, 123), (325, 45)]" + } + ] + }, + { + "id": 566, + "image": [ + "images/566_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the rag\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(327, 94), (254, 129), (183, 168), (112, 214)]" + } + ] + }, + { + "id": 33174307, + "image": [ + "images/33174307_15.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Stack the red cube on top of the green cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(128, 128), (122, 136), (124, 140), (133, 136), (143, 147)]" + } + ] + }, + { + "id": 3793, + "image": [ + "images/3793_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(113, 253), (103, 208), (99, 161), (107, 139)]" + } + ] + }, + { + "id": 25514701, + "image": [ + "images/25514701_108.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place the pepper in the ceramic bowl. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(66, 86), (67, 90), (63, 86), (92, 69), (105, 64)]" + } + ] + }, + { + "id": 11555052, + "image": [ + "images/11555052_564.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(180, 84), (181, 84), (181, 84), (178, 86), (176, 81)]" + } + ] + }, + { + "id": 17948694, + "image": [ + "images/17948694_184.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is open bottom drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(166, 207), (168, 211), (171, 207), (168, 219), (170, 215)]" + } + ] + }, + { + "id": 18332132, + "image": [ + "images/18332132_2.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is move orange near 7up can. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(288, 81), (197, 69), (192, 79), (128, 123), (221, 68)]" + } + ] + }, + { + "id": 4577, + "image": [ + "images/4577_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(120, 111), (161, 111), (198, 113), (242, 115), (281, 120), (335, 120)]" + } + ] + }, + { + "id": 27018718, + "image": [ + "images/27018718_137.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with sponge. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(80, 90), (73, 95), (69, 103), (70, 109), (67, 114)]" + } + ] + }, + { + "id": 3026, + "image": [ + "images/3026_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mushroom in the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(195, 99), (241, 105), (289, 118), (341, 164)]" + } + ] + }, + { + "id": 18237860, + "image": [ + "images/18237860_16.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is pick blue plastic bottle. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(85, 119), (89, 119), (80, 108), (67, 89), (64, 82)]" + } + ] + }, + { + "id": 22723644, + "image": [ + "images/22723644_2.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(472, 302), (392, 330), (376, 342), (284, 290), (222, 308)]" + } + ] + }, + { + "id": 4268, + "image": [ + "images/4268_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the button\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(118, 209), (113, 186), (112, 161)]" + } + ] + }, + { + "id": 4325508, + "image": [ + "images/4325508_0162.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(62, 196), (55, 170), (224, 168), (437, 264), (478, 44)]" + } + ] + }, + { + "id": 15858295, + "image": [ + "images/15858295_10.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is move the blue triangle towards the yellow hexagon. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(195, 147), (179, 117), (273, 146), (362, 120)]" + } + ] + }, + { + "id": 2, + "image": [ + "images/2_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the croissant\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(104, 33), (97, 39), (92, 50), (88, 59)]" + } + ] + }, + { + "id": 6037, + "image": [ + "images/6037_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(131, 114), (181, 120), (223, 136), (270, 154)]" + } + ] + }, + { + "id": 11372736, + "image": [ + "images/11372736_230.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(178, 79), (180, 81), (181, 86), (181, 79), (176, 81)]" + } + ] + }, + { + "id": 11720865, + "image": [ + "images/11720865_406.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(187, 101), (191, 109), (180, 101), (196, 109), (200, 96)]" + } + ] + }, + { + "id": 10443781, + "image": [ + "images/10443781_149.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(183, 99), (180, 99), (183, 99), (189, 99), (180, 99)]" + } + ] + }, + { + "id": 5129, + "image": [ + "images/5129_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the strainer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(103, 116), (117, 103), (131, 102), (159, 102)]" + } + ] + }, + { + "id": 5970, + "image": [ + "images/5970_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(119, 111), (164, 118), (203, 125), (242, 139)]" + } + ] + }, + { + "id": 2954, + "image": [ + "images/2954_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(248, 96), (218, 237), (241, 277)]" + } + ] + }, + { + "id": 13473464, + "image": [ + "images/13473464_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a WidowX robot using the end effector control. The task is take carrot off plate cardboardfence. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(209, 172), (186, 244), (156, 280), (224, 64), (170, 98)]" + } + ] + }, + { + "id": 9103065, + "image": [ + "images/9103065_0385.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick red green blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(300, 149), (302, 144), (302, 144), (300, 140), (300, 140)]" + } + ] + }, + { + "id": 6143, + "image": [ + "images/6143_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(125, 118), (180, 109), (228, 94), (270, 70)]" + } + ] + }, + { + "id": 17059094, + "image": [ + "images/17059094_15.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is slide green circle on top of red circle and hexagon. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(341, 51), (328, 36), (247, 41), (282, 39), (284, 33)]" + } + ] + }, + { + "id": 6192, + "image": [ + "images/6192_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(111, 112), (150, 90), (184, 82), (222, 79)]" + } + ] + }, + { + "id": 1836, + "image": [ + "images/1836_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(252, 170), (288, 187), (304, 211), (304, 247)]" + } + ] + }, + { + "id": 6198, + "image": [ + "images/6198_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the banana on the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(116, 114), (167, 130), (208, 132), (261, 130)]" + } + ] + }, + { + "id": 11405107, + "image": [ + "images/11405107_51.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(131, 127), (176, 68), (168, 109), (172, 122), (196, 114)]" + } + ] + }, + { + "id": 5719, + "image": [ + "images/5719_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(156, 116), (169, 128), (175, 136), (178, 150)]" + } + ] + }, + { + "id": 14718878, + "image": [ + "images/14718878_24.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is place the yellow heart horizontally left to the red circle. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(188, 99), (218, 105), (197, 105), (256, 77), (259, 65)]" + } + ] + }, + { + "id": 2271, + "image": [ + "images/2271_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"lower the gripper towards the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(97, 213), (174, 222), (248, 228), (328, 250), (398, 302)]" + } + ] + }, + { + "id": 6046, + "image": [ + "images/6046_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle in the strainer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(131, 108), (139, 126), (153, 144), (170, 155)]" + } + ] + }, + { + "id": 809, + "image": [ + "images/809_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the pan\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(231, 117), (312, 126), (373, 134), (488, 223)]" + } + ] + }, + { + "id": 28268704, + "image": [ + "images/28268704_90.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place the eraser on the white sponge. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(103, 70), (105, 70), (103, 70), (105, 68), (103, 69)]" + } + ] + }, + { + "id": 3549, + "image": [ + "images/3549_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the object in the first container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(371, 223), (289, 244), (261, 259)]" + } + ] + }, + { + "id": 21487705, + "image": [ + "images/21487705_5.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(437, 356), (433, 306), (432, 332), (440, 335), (460, 240)]" + } + ] + }, + { + "id": 21948875, + "image": [ + "images/21948875_15.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(252, 257), (250, 251), (241, 306), (273, 172)]" + } + ] + }, + { + "id": 4613, + "image": [ + "images/4613_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(142, 109), (164, 121), (198, 128), (239, 132), (262, 136)]" + } + ] + }, + { + "id": 971, + "image": [ + "images/971_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the can\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(277, 226), (382, 245), (432, 212)]" + } + ] + }, + { + "id": 15047872, + "image": [ + "images/15047872_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is move the blue triangle towards the top left. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(323, 23), (254, 23), (231, 18), (195, 18), (195, 18)]" + } + ] + }, + { + "id": 2078, + "image": [ + "images/2078_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(64, 226), (161, 204), (227, 184), (252, 176)]" + } + ] + }, + { + "id": 6823, + "image": [ + "images/6823_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(116, 109), (158, 109), (208, 115), (245, 128)]" + } + ] + }, + { + "id": 6666, + "image": [ + "images/6666_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(131, 119), (145, 132), (156, 139), (170, 149)]" + } + ] + }, + { + "id": 12242369, + "image": [ + "images/12242369_822.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(172, 75), (69, 66), (165, 34), (170, 49), (165, 45)]" + } + ] + }, + { + "id": 17456351, + "image": [ + "images/17456351_46.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is move paper bowl near apple. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(160, 11), (161, 11), (169, 17), (166, 35), (163, 33)]" + } + ] + }, + { + "id": 26219537, + "image": [ + "images/26219537_136.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with eraser. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(79, 120), (79, 120), (80, 120), (79, 120), (79, 121)]" + } + ] + }, + { + "id": 18829288, + "image": [ + "images/18829288_48.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is open top drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(206, 152), (193, 150), (193, 167), (211, 179), (230, 178)]" + } + ] + }, + { + "id": 4026, + "image": [ + "images/4026_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the button\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(115, 206), (120, 133), (125, 125)]" + } + ] + }, + { + "id": 1687, + "image": [ + "images/1687_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the potato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(293, 139), (309, 219), (316, 321), (320, 410)]" + } + ] + }, + { + "id": 24148424, + "image": [ + "images/24148424_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(353, 334), (352, 353), (375, 292), (376, 302), (353, 366)]" + } + ] + }, + { + "id": 14189282, + "image": [ + "images/14189282_16.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is push the blue triangle towards the yellow heart. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(476, 42), (467, 38), (465, 38), (366, 45)]" + } + ] + }, + { + "id": 25596583, + "image": [ + "images/25596583_130.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place pepper in metal cup. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(70, 99), (61, 97), (59, 101), (69, 87), (70, 86)]" + } + ] + }, + { + "id": 364, + "image": [ + "images/364_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"pick up the green cup\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(189, 8), (168, 12), (129, 17), (105, 21), (86, 28), (45, 40), (30, 42)]" + } + ] + }, + { + "id": 27685269, + "image": [ + "images/27685269_86.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place towel in table surface. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(94, 75), (94, 75), (94, 75), (94, 74), (94, 75)]" + } + ] + }, + { + "id": 3505702, + "image": [ + "images/3505702_0408.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(414, 132), (410, 219), (410, 232), (432, 60), (440, 34)]" + } + ] + }, + { + "id": 6298, + "image": [ + "images/6298_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(113, 112), (169, 138), (203, 148), (233, 130)]" + } + ] + }, + { + "id": 796755, + "image": [ + "images/796755_1465.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Interact with the objects in diverse but meaningful ways. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(40, 44), (70, 8), (24, 24), (70, 9), (72, 13)]" + } + ] + }, + { + "id": 599, + "image": [ + "images/599_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(85, 189), (163, 211), (240, 241), (337, 309)]" + } + ] + }, + { + "id": 5234, + "image": [ + "images/5234_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"pick up the lid of the tea container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(214, 4), (250, 17), (292, 43), (312, 66)]" + } + ] + }, + { + "id": 4016, + "image": [ + "images/4016_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the cabinet door handles\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(120, 208), (144, 113), (136, 102)]" + } + ] + }, + { + "id": 1325, + "image": [ + "images/1325_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the stuffed animal\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(350, 15), (394, 107), (448, 205), (467, 310)]" + } + ] + }, + { + "id": 3912066, + "image": [ + "images/3912066_0006.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is destack yellow blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(336, 43), (300, 216), (323, 181), (396, 373), (416, 116)]" + } + ] + }, + { + "id": 20373004, + "image": [ + "images/20373004_3.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is open middle drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(281, 115), (240, 214), (238, 214), (254, 221), (262, 216)]" + } + ] + }, + { + "id": 5045, + "image": [ + "images/5045_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(106, 106), (161, 125), (226, 130), (292, 120)]" + } + ] + }, + { + "id": 5343, + "image": [ + "images/5343_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(137, 115), (215, 134), (256, 134)]" + } + ] + }, + { + "id": 12495911, + "image": [ + "images/12495911_185.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble lamp. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(163, 42), (172, 45), (170, 47), (71, 36), (159, 19)]" + } + ] + }, + { + "id": 14160160, + "image": [ + "images/14160160_11.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is push the red circle and green circle to the left side of the board. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(296, 24), (241, 30), (245, 24), (261, 20), (256, 28)]" + } + ] + }, + { + "id": 4532, + "image": [ + "images/4532_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(136, 108), (175, 121), (220, 128), (272, 132), (293, 136)]" + } + ] + }, + { + "id": 1759160, + "image": [ + "images/1759160_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(133, 25), (208, 36), (178, 63), (189, 60), (170, 70)]" + } + ] + }, + { + "id": 5601041, + "image": [ + "images/5601041_0114.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(444, 103), (432, 175), (426, 194), (449, 82), (451, 43)]" + } + ] + }, + { + "id": 12710602, + "image": [ + "images/12710602_85.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(178, 90), (168, 68), (187, 103), (174, 116), (198, 140)]" + } + ] + }, + { + "id": 20396683, + "image": [ + "images/20396683_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is open middle drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(281, 187), (224, 192), (224, 189), (226, 211), (240, 199)]" + } + ] + }, + { + "id": 30235189, + "image": [ + "images/30235189_231.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Turn on the faucet by rotating a designated handle. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(182, 56), (184, 56), (186, 57), (188, 55)]" + } + ] + }, + { + "id": 6528, + "image": [ + "images/6528_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"pull the strainer backwards\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(159, 131), (155, 150), (156, 178), (164, 194)]" + } + ] + }, + { + "id": 24626661, + "image": [ + "images/24626661_2.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(407, 291), (421, 300), (396, 328), (423, 300), (476, 243)]" + } + ] + }, + { + "id": 781, + "image": [ + "images/781_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the bowl\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(129, 103), (188, 134), (227, 175), (245, 252)]" + } + ] + }, + { + "id": 5496, + "image": [ + "images/5496_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move the mug to the right\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(198, 148), (222, 149), (251, 144), (268, 138)]" + } + ] + }, + { + "id": 2561, + "image": [ + "images/2561_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move the square above the fork\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(177, 208), (211, 187), (268, 180), (311, 187)]" + } + ] + }, + { + "id": 32148539, + "image": [ + "images/32148539_73.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(109, 122), (110, 128), (109, 129), (112, 118), (114, 109)]" + } + ] + }, + { + "id": 25926531, + "image": [ + "images/25926531_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is drag the pepper across the table. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(135, 62), (122, 63), (102, 82), (90, 97), (48, 99)]" + } + ] + }, + { + "id": 587, + "image": [ + "images/587_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the cloth\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(444, 197), (360, 177), (282, 160), (243, 163)]" + } + ] + }, + { + "id": 17651079, + "image": [ + "images/17651079_31.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is open top drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(148, 218), (148, 224), (149, 228), (154, 231)]" + } + ] + }, + { + "id": 7842628, + "image": [ + "images/7842628_0091.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack green blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(295, 173), (270, 348), (323, 211), (380, 216), (433, 26)]" + } + ] + }, + { + "id": 9980701, + "image": [ + "images/9980701_119.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(180, 68), (105, 86), (176, 69), (176, 62), (202, 68)]" + } + ] + }, + { + "id": 12998999, + "image": [ + "images/12998999_45.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Insert the blue gear onto the right peg, followed by the red gear. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(63, 13), (59, 15), (47, 24), (47, 6), (44, 26)]" + } + ] + }, + { + "id": 24256186, + "image": [ + "images/24256186_5.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(483, 352), (485, 352), (483, 351), (476, 329), (467, 348)]" + } + ] + }, + { + "id": 2158, + "image": [ + "images/2158_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"place the banana next to the bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(177, 99), (213, 173), (238, 248), (272, 311)]" + } + ] + }, + { + "id": 5142, + "image": [ + "images/5142_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(133, 114), (144, 134), (164, 147), (192, 149)]" + } + ] + }, + { + "id": 15979875, + "image": [ + "images/15979875_11.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is push yellow hexagon along with the green circle to the right side of the board. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(240, 128), (305, 137), (343, 137), (419, 136)]" + } + ] + }, + { + "id": 12017587, + "image": [ + "images/12017587_481.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(191, 71), (189, 71), (187, 75), (187, 73), (185, 73)]" + } + ] + }, + { + "id": 323, + "image": [ + "images/323_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"lift the sponge from the blue container\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(211, 77), (195, 60), (170, 56), (133, 68)]" + } + ] + }, + { + "id": 24169703, + "image": [ + "images/24169703_6.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(327, 309), (330, 290), (364, 249), (348, 290), (311, 306)]" + } + ] + }, + { + "id": 730810, + "image": [ + "images/730810_885.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Interact with the objects in diverse but meaningful ways. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(71, 16), (107, 34), (64, 19), (29, 86), (27, 93)]" + } + ] + }, + { + "id": 32462209, + "image": [ + "images/32462209_105.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Insert the peg into the horizontal hole in a box. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(124, 96), (125, 93), (128, 96), (130, 91), (130, 90)]" + } + ] + }, + { + "id": 17584409, + "image": [ + "images/17584409_23.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is place 7up can upright. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(196, 64), (200, 67), (201, 69), (214, 81), (232, 88)]" + } + ] + }, + { + "id": 6252498, + "image": [ + "images/6252498_0205.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack gray white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(272, 80), (273, 156), (275, 144), (273, 44), (277, 29)]" + } + ] + }, + { + "id": 28418885, + "image": [ + "images/28418885_74.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is wipe red bowl with eraser. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(165, 100), (116, 80), (119, 81), (119, 81), (130, 86)]" + } + ] + }, + { + "id": 2023203, + "image": [ + "images/2023203_21.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(201, 37), (186, 37), (179, 36), (177, 42), (181, 33)]" + } + ] + }, + { + "id": 4793, + "image": [ + "images/4793_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(114, 132), (195, 102), (242, 81), (272, 61)]" + } + ] + }, + { + "id": 1959009, + "image": [ + "images/1959009_19.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(96, 109), (81, 116), (129, 111), (173, 144), (176, 121)]" + } + ] + }, + { + "id": 12213000, + "image": [ + "images/12213000_1141.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble square_table. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(133, 116), (170, 79), (180, 75), (112, 75), (103, 71)]" + } + ] + }, + { + "id": 3112, + "image": [ + "images/3112_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move the corn towards the middle right of the table\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(590, 79), (554, 97), (519, 132), (494, 176), (485, 213), (485, 261), (485, 294)]" + } + ] + }, + { + "id": 28220565, + "image": [ + "images/28220565_79.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place pepper in purple bowl. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(145, 91), (145, 90), (135, 74), (122, 70), (122, 71)]" + } + ] + }, + { + "id": 16581704, + "image": [ + "images/16581704_24.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is move the yellow hexagon diagonally right to the green circle. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(378, 212), (371, 208), (378, 208), (380, 189), (405, 182)]" + } + ] + }, + { + "id": 31019165, + "image": [ + "images/31019165_60.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(109, 154), (108, 154), (115, 150), (168, 119), (192, 103)]" + } + ] + }, + { + "id": 2131, + "image": [ + "images/2131_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the sushi roll\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(321, 17), (302, 84), (282, 140), (266, 204)]" + } + ] + }, + { + "id": 25145241, + "image": [ + "images/25145241_90.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place eraser in tray. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(119, 82), (116, 58), (112, 53), (118, 66), (119, 64)]" + } + ] + }, + { + "id": 30600931, + "image": [ + "images/30600931_18.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(123, 124), (122, 143), (122, 154), (121, 132), (123, 43)]" + } + ] + }, + { + "id": 5836113, + "image": [ + "images/5836113_0211.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack boston blue white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(419, 340), (403, 240), (289, 164), (188, 144), (176, 24)]" + } + ] + }, + { + "id": 5759626, + "image": [ + "images/5759626_0256.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack purple white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(97, 202), (163, 153), (307, 142), (336, 208), (343, 34)]" + } + ] + }, + { + "id": 2363, + "image": [ + "images/2363_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the banana in the colander\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(291, 103), (328, 136), (352, 189), (360, 245), (373, 342)]" + } + ] + }, + { + "id": 28833023, + "image": [ + "images/28833023_36.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place sponge in tray. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(69, 77), (50, 95), (48, 101), (51, 104), (51, 107)]" + } + ] + }, + { + "id": 8810231, + "image": [ + "images/8810231_0157.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick red green cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(392, 245), (394, 210), (396, 134), (405, 54), (407, 46)]" + } + ] + }, + { + "id": 430, + "image": [ + "images/430_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move the bread towards the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(77, 42), (64, 38), (53, 40), (40, 45)]" + } + ] + }, + { + "id": 3365, + "image": [ + "images/3365_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the yellow brush\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(248, 142), (273, 184), (302, 236), (293, 286)]" + } + ] + }, + { + "id": 4231, + "image": [ + "images/4231_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the cabinet door\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(123, 204), (123, 184), (112, 164), (101, 152), (79, 132)]" + } + ] + }, + { + "id": 138, + "image": [ + "images/138_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(36, 36), (43, 49), (51, 58), (66, 68)]" + } + ] + }, + { + "id": 7288602, + "image": [ + "images/7288602_0020.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick red green blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(293, 39), (227, 85), (216, 162), (220, 190), (211, 115)]" + } + ] + }, + { + "id": 2479394, + "image": [ + "images/2479394_25.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(171, 64), (182, 58), (190, 58), (197, 48), (198, 50)]" + } + ] + }, + { + "id": 5930764, + "image": [ + "images/5930764_0174.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(316, 197), (320, 168), (321, 115), (325, 85), (328, 38)]" + } + ] + }, + { + "id": 4275, + "image": [ + "images/4275_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the drawer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(110, 198), (119, 163), (125, 130), (135, 87)]" + } + ] + }, + { + "id": 22157960, + "image": [ + "images/22157960_11.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(405, 311), (364, 329), (353, 352), (357, 339), (437, 244)]" + } + ] + }, + { + "id": 5894, + "image": [ + "images/5894_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(113, 108), (128, 131), (145, 150), (170, 165)]" + } + ] + }, + { + "id": 28570949, + "image": [ + "images/28570949_29.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is wipe table surface with towel. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(172, 104), (142, 122), (133, 121), (119, 110), (122, 112)]" + } + ] + }, + { + "id": 639, + "image": [ + "images/639_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(96, 49), (138, 81), (184, 121), (229, 155)]" + } + ] + }, + { + "id": 4707, + "image": [ + "images/4707_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move the mug to the right\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(212, 147), (231, 139), (250, 144), (262, 157)]" + } + ] + }, + { + "id": 4197, + "image": [ + "images/4197_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the microwave oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(107, 215), (141, 165), (174, 98)]" + } + ] + }, + { + "id": 6803675, + "image": [ + "images/6803675_0229.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is destack boston blue gray cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(136, 111), (197, 157), (296, 257), (323, 310), (323, 106)]" + } + ] + }, + { + "id": 2764, + "image": [ + "images/2764_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"place the apple on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(179, 217), (188, 151), (186, 96), (177, 32)]" + } + ] + }, + { + "id": 1323, + "image": [ + "images/1323_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the spatula\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(142, 56), (193, 78), (248, 142), (272, 252)]" + } + ] + }, + { + "id": 31323233, + "image": [ + "images/31323233_61.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Insert a designated object into the corresponding slot on a board. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(162, 88), (152, 90), (93, 105), (52, 120), (55, 122)]" + } + ] + }, + { + "id": 13778353, + "image": [ + "images/13778353_0.jpg" + ], + "conversations": [ + { + "value": "\n You are a WidowX robot using the end effector control. The task is move towel to lower right corner. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(256, 110), (362, 147), (392, 173), (435, 303), (451, 245)]" + } + ] + }, + { + "id": 1512170, + "image": [ + "images/1512170_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(105, 94), (114, 96), (104, 66), (96, 63), (141, 61)]" + } + ] + }, + { + "id": 1153, + "image": [ + "images/1153_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the tomato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(277, 105), (279, 128), (270, 152), (245, 96)]" + } + ] + }, + { + "id": 12633235, + "image": [ + "images/12633235_252.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble one_leg. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(12, 152), (12, 152), (12, 152), (185, 105), (12, 152)]" + } + ] + }, + { + "id": 15944740, + "image": [ + "images/15944740_8.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is separate the blue blocks. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(49, 87), (128, 182), (243, 267), (464, 244), (483, 227)]" + } + ] + }, + { + "id": 4746, + "image": [ + "images/4746_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(144, 119), (178, 116), (220, 116), (251, 120)]" + } + ] + }, + { + "id": 8171470, + "image": [ + "images/8171470_0129.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick 4wp wood cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(229, 170), (229, 202), (231, 204), (232, 156), (225, 49)]" + } + ] + }, + { + "id": 8022683, + "image": [ + "images/8022683_0084.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack green blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(250, 139), (213, 254), (272, 124), (296, 181), (300, 15)]" + } + ] + }, + { + "id": 2512, + "image": [ + "images/2512_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the spoon\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(259, 48), (199, 85), (133, 146), (85, 216)]" + } + ] + }, + { + "id": 23048, + "image": [ + "images/23048_74.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Kindly get your arm to the topmost drawer handle.\nGet a good grip on it, then softly pull to open it. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(86, 34), (102, 29), (104, 32), (96, 35), (91, 39)]" + } + ] + }, + { + "id": 10559325, + "image": [ + "images/10559325_1694.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble chair. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(114, 34), (107, 34), (112, 32), (96, 36), (98, 36)]" + } + ] + }, + { + "id": 4787, + "image": [ + "images/4787_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(130, 114), (172, 116), (206, 119), (257, 130)]" + } + ] + }, + { + "id": 357, + "image": [ + "images/357_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"pick up the breadstick\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(146, 122), (133, 99), (111, 81), (88, 71), (66, 62), (45, 60)]" + } + ] + }, + { + "id": 3464533, + "image": [ + "images/3464533_0047.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(234, 84), (103, 169), (83, 240), (71, 192), (44, 84)]" + } + ] + }, + { + "id": 29094723, + "image": [ + "images/29094723_72.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is place towel in plastic cup. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(57, 75), (51, 68), (56, 71), (56, 73), (57, 70)]" + } + ] + }, + { + "id": 3959, + "image": [ + "images/3959_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"approach the oven\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(130, 227), (133, 194), (132, 173), (130, 140), (124, 115), (115, 90)]" + } + ] + }, + { + "id": 22146877, + "image": [ + "images/22146877_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(476, 225), (481, 244), (492, 242), (471, 235), (401, 269)]" + } + ] + }, + { + "id": 6521133, + "image": [ + "images/6521133_0242.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is stack gray white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(204, 105), (206, 144), (204, 108), (195, 41), (241, 12)]" + } + ] + }, + { + "id": 5155, + "image": [ + "images/5155_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the ketchup bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(156, 160), (178, 144), (206, 136), (236, 130)]" + } + ] + }, + { + "id": 1124, + "image": [ + "images/1124_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the cabinet handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(528, 256), (512, 297), (476, 328), (439, 347)]" + } + ] + }, + { + "id": 4445, + "image": [ + "images/4445_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"hover over the round object\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(123, 3), (122, 15), (119, 33), (119, 48)]" + } + ] + }, + { + "id": 14248106, + "image": [ + "images/14248106_11.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is push the red circle to the top of the yellow hexagon. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(74, 55), (368, 50), (449, 39), (464, 55), (412, 68)]" + } + ] + }, + { + "id": 20796777, + "image": [ + "images/20796777_13.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(563, 239), (545, 307), (533, 227), (524, 305), (540, 309)]" + } + ] + }, + { + "id": 15124148, + "image": [ + "images/15124148_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is move the green star into the green circle. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(129, 50), (154, 50), (163, 48), (167, 51), (145, 47)]" + } + ] + }, + { + "id": 6186, + "image": [ + "images/6186_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(114, 114), (192, 107), (262, 114), (348, 134)]" + } + ] + }, + { + "id": 6444, + "image": [ + "images/6444_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the bottle on the plate\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(155, 116), (203, 105), (231, 78)]" + } + ] + }, + { + "id": 528, + "image": [ + "images/528_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the bowl\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(426, 158), (385, 189), (359, 213), (328, 241)]" + } + ] + }, + { + "id": 21124308, + "image": [ + "images/21124308_11.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(456, 332), (417, 365), (432, 352), (400, 392), (405, 357)]" + } + ] + }, + { + "id": 12641200, + "image": [ + "images/12641200_545.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble desk. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(73, 51), (118, 75), (176, 62), (73, 58), (77, 71)]" + } + ] + }, + { + "id": 6205, + "image": [ + "images/6205_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(117, 118), (176, 120), (228, 132), (267, 152)]" + } + ] + }, + { + "id": 10348867, + "image": [ + "images/10348867_199.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is assemble stool. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(172, 64), (79, 101), (152, 51), (150, 53), (142, 64)]" + } + ] + }, + { + "id": 21975924, + "image": [ + "images/21975924_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(506, 311), (506, 314), (508, 314), (506, 314), (508, 312)]" + } + ] + }, + { + "id": 8764635, + "image": [ + "images/8764635_0255.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is destack red blue cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(385, 340), (378, 360), (385, 319), (389, 206), (400, 156)]" + } + ] + }, + { + "id": 6337, + "image": [ + "images/6337_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(128, 112), (175, 138), (223, 143), (265, 138)]" + } + ] + }, + { + "id": 1555635, + "image": [ + "images/1555635_13.jpg" + ], + "conversations": [ + { + "value": "\n You are a Multi Robot robot using the end effector control. The task is Interact with the objects in the bin. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(230, 70), (210, 68), (194, 64), (225, 54), (214, 84)]" + } + ] + }, + { + "id": 675, + "image": [ + "images/675_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"grip the vessel\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(213, 76), (257, 164), (291, 228), (328, 276)]" + } + ] + }, + { + "id": 17598097, + "image": [ + "images/17598097_23.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is pick orange from white bowl. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(165, 65), (189, 50), (214, 50), (216, 79)]" + } + ] + }, + { + "id": 28501760, + "image": [ + "images/28501760_2.jpg" + ], + "conversations": [ + { + "value": "\n You are a Google Robot robot using the end effector control. The task is wipe tray with sponge. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(63, 73), (63, 74), (60, 69), (44, 69), (51, 70)]" + } + ] + }, + { + "id": 5111, + "image": [ + "images/5111_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the mug\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(142, 114), (173, 121), (203, 127), (251, 139)]" + } + ] + }, + { + "id": 20068889, + "image": [ + "images/20068889_12.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is move green can near redbull can. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(144, 45), (122, 45), (161, 78), (115, 120), (234, 93)]" + } + ] + }, + { + "id": 4819, + "image": [ + "images/4819_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"drag the mug forward\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(262, 114), (270, 83), (259, 58), (262, 33)]" + } + ] + }, + { + "id": 1927, + "image": [ + "images/1927_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the potato\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(55, 84), (83, 117), (113, 130), (172, 128)]" + } + ] + }, + { + "id": 4281, + "image": [ + "images/4281_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the handle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(108, 215), (119, 173), (119, 128), (119, 105)]" + } + ] + }, + { + "id": 18398623, + "image": [ + "images/18398623_1.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is pick orange can from middle drawer and place on counter. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(296, 88), (234, 187), (192, 189), (216, 48), (222, 58)]" + } + ] + }, + { + "id": 7602610, + "image": [ + "images/7602610_0221.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is destack gray white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(170, 204), (170, 202), (165, 156), (154, 73), (152, 48)]" + } + ] + }, + { + "id": 6012, + "image": [ + "images/6012_frame_15.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"close the drawer\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(181, 93), (170, 102), (161, 118), (151, 130)]" + } + ] + }, + { + "id": 143, + "image": [ + "images/143_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"move towards the red object\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(43, 96), (68, 103), (83, 111), (96, 118), (99, 133)]" + } + ] + }, + { + "id": 5440609, + "image": [ + "images/5440609_0241.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the joint control. The task is pick gray white cube. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(353, 45), (353, 46), (352, 46), (353, 46), (352, 46)]" + } + ] + }, + { + "id": 15039374, + "image": [ + "images/15039374_10.jpg" + ], + "conversations": [ + { + "value": "\n You are a xArm robot using the end effector control. The task is nudge the red circle right a bit. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(298, 20), (296, 30), (323, 20), (346, 29)]" + } + ] + }, + { + "id": 33190562, + "image": [ + "images/33190562_9.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(122, 121), (121, 124), (122, 135), (122, 135), (98, 61)]" + } + ] + }, + { + "id": 21153184, + "image": [ + "images/21153184_13.jpg" + ], + "conversations": [ + { + "value": "\n You are a Kuka Iiwa robot using the end effector control. The task is pick anything. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(508, 280), (503, 303), (499, 311), (533, 345), (483, 390)]" + } + ] + }, + { + "id": 19262395, + "image": [ + "images/19262395_7.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is close middle drawer. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(224, 224), (216, 226), (201, 235), (202, 194), (208, 186)]" + } + ] + }, + { + "id": 5012, + "image": [ + "images/5012_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(150, 124), (228, 121), (251, 138)]" + } + ] + }, + { + "id": 3224, + "image": [ + "images/3224_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the pot\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(147, 66), (206, 118), (270, 180), (293, 267)]" + } + ] + }, + { + "id": 5928, + "image": [ + "images/5928_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the banana\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(109, 112), (153, 121), (200, 130), (251, 132)]" + } + ] + }, + { + "id": 32557112, + "image": [ + "images/32557112_110.jpg" + ], + "conversations": [ + { + "value": "\n You are a Franka robot using the end effector control. The task is Pick up the object and move it to a goal position. Please predict up to 10 key trajectory points to complete the task.", + "from": "human" + }, + { + "from": "gpt", + "value": "[(115, 104), (108, 84), (96, 60), (88, 38), (80, 37)]" + } + ] + }, + { + "id": 3868, + "image": [ + "images/3868_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"reach for the bottle\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(123, 174), (99, 115), (116, 81)]" + } + ] + }, + { + "id": 543, + "image": [ + "images/543_frame_0.png" + ], + "conversations": [ + { + "from": "human", + "value": "\nYou are a robot using the joint control. The task is \"pick up the pan\". Please predict up to 10 key trajectory points to complete the task. Your answer should be formatted as a list of tuples, i.e. [[x1, y1], [x2, y2], ...], where each tuple contains the x and y coordinates of a point." + }, + { + "from": "gpt", + "value": "[(206, 48), (211, 127), (225, 215), (248, 292)]" + } + ] + } ] \ No newline at end of file