Dataset Viewer
Auto-converted to Parquet
id
string
image_src
image
image_tgt
image
mask
image
point
list
object
string
prompt_change
string
prompt_mask
string
prompt_point
string
crossview
string
split
string
tr000000_obj0
[ [ 0.47749999165534973, 0.9661654233932495 ] ]
altar
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000000_obj1
[ [ 0.6850000023841858, 0.5751879811286926 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000000_obj2
[ [ 0.47999998927116394, 0.37593984603881836 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000000_obj3
[ [ 0.26499998569488525, 0.5939849615097046 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000001_obj0
[ [ 0.5944206118583679, 0.5596638917922974 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000001_obj1
[ [ 0.2596566379070282, 0.2655462324619293 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000001_obj2
[ [ 0.9527897238731384, 0.48235294222831726 ] ]
column
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000001_obj3
[ [ 0.9399141669273376, 0.15126051008701324 ] ]
capital
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000002_obj0
[ [ 0.7875000238418579, 0.7118644118309021 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000002_obj1
[ [ 0.4950000047683716, 0.9220339059829712 ] ]
cross
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000002_obj2
[ [ 0.19499999284744263, 0.7152542471885681 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000003_obj0
[ [ 0.5185185074806213, 0.957446813583374 ] ]
stairway
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000003_obj1
[ [ 0.5502645373344421, 0.8936170339584351 ] ]
altar
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000003_obj2
[ [ 0.7407407164573669, 0.7234042286872864 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000003_obj3
[ [ 0.2063492089509964, 0.7191489338874817 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000004_obj0
[ [ 0.49000000953674316, 0.8907216787338257 ] ]
altar
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000004_obj1
[ [ 0.6740000247955322, 0.7360824942588806 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000004_obj2
[ [ 0.45399999618530273, 0.9587628841400146 ] ]
stairway
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000004_obj3
[ [ 0.3479999899864197, 0.7298969030380249 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000005_obj0
[ [ 0.07428571581840515, 0.8626609444618225 ] ]
bench
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000005_obj1
[ [ 0.5028571486473083, 0.28326180577278137 ] ]
vault
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000005_obj2
[ [ 0.10857142508029938, 0.05150214582681656 ] ]
lamp
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000005_obj3
[ [ 0.22857142984867096, 0.7553648352622986 ] ]
railing
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000006_obj0
[ [ 0.875, 0.7360000014305115 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000006_obj1
[ [ 0.5149999856948853, 0.7400000095367432 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000006_obj2
[ [ 0.1550000011920929, 0.7279999852180481 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000007_obj0
[ [ 0.04500000178813934, 0.7955555319786072 ] ]
mezzanine
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000007_obj1
[ [ 0.9166666865348816, 0.7400000095367432 ] ]
mezzanine
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000007_obj2
[ [ 0.878333330154419, 0.695555567741394 ] ]
door
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000008_obj0
[ [ 0.4350000023841858, 0.7133333086967468 ] ]
cross
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000008_obj1
[ [ 0.7774999737739563, 0.5666666626930237 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000008_obj2
[ [ 0.5699999928474426, 0.6000000238418579 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000008_obj3
[ [ 0.27000001072883606, 0.476666659116745 ] ]
windowpane
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000009_obj0
[ [ 0.6899999976158142, 0.7833572626113892 ] ]
column
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000009_obj1
[ [ 0.36375001072883606, 0.8048780560493469 ] ]
candelabrum
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000009_obj2
[ [ 0.23375000059604645, 0.8134863972663879 ] ]
column
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000009_obj3
[ [ 0.7724999785423279, 0.9583930969238281 ] ]
bench
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000010_obj0
[ [ 0.5192307829856873, 0.39587628841400146 ] ]
head
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000010_obj1
[ [ 0.776098906993866, 0.5752577185630798 ] ]
left foot
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000010_obj2
[ [ 0.8928571343421936, 0.14432989060878754 ] ]
mouth
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000010_obj3
[ [ 0.8626373410224915, 0.5793814659118652 ] ]
left leg
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000011_obj0
[ [ 0.9380000233650208, 0.7120000123977661 ] ]
rock
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000011_obj1
[ [ 0.33399999141693115, 0.874666690826416 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000011_obj2
[ [ 0.6880000233650208, 0.9066666960716248 ] ]
fence
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000011_obj3
[ [ 0.10999999940395355, 0.8426666855812073 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000012_obj0
[ [ 0.6416666507720947, 0.6800000071525574 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000012_obj1
[ [ 0.7950000166893005, 0.1666666716337204 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000012_obj2
[ [ 0.8916666507720947, 0.5533333420753479 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000012_obj3
[ [ 0.49666666984558105, 0.33666667342185974 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000013_obj0
[ [ 0.3634146451950073, 0.6262975931167603 ] ]
gate
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000013_obj1
[ [ 0.5, 0.6435986161231995 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000013_obj2
[ [ 0.43414634466171265, 0.6401383876800537 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000013_obj3
[ [ 0.5658536553382874, 0.6782007217407227 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000014_obj0
[ [ 0.699999988079071, 0.7875000238418579 ] ]
left hand
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000014_obj1
[ [ 0.671875, 0.9520833492279053 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000014_obj2
[ [ 0.5171874761581421, 0.75 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000014_obj3
[ [ 0.609375, 0.5625 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000015_obj0
[ [ 0.6449999809265137, 0.05607476457953453 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000015_obj1
[ [ 0.5149999856948853, 0.2841121554374695 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000015_obj2
[ [ 0.4325000047683716, 0.46355140209198 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000015_obj3
[ [ 0.6575000286102295, 0.5607476830482483 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000016_obj0
[ [ 0.05999999865889549, 0.0560000017285347 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000016_obj1
[ [ 0.3540000021457672, 0.08266666531562805 ] ]
rock
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000016_obj2
[ [ 0.7459999918937683, 0.4880000054836273 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000017_obj0
[ [ 0.6725000143051147, 0.8986867070198059 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000017_obj1
[ [ 0.38749998807907104, 0.8968105316162109 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000017_obj2
[ [ 0.061250001192092896, 0.904315173625946 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000017_obj3
[ [ 0.6650000214576721, 0.35084426403045654 ] ]
shark
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000018_obj0
[ [ 0.3840000033378601, 0.949999988079071 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000018_obj1
[ [ 0.5239999890327454, 0.203125 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000018_obj2
[ [ 0.5600000023841858, 0.953125 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000018_obj3
[ [ 0.6340000033378601, 0.940625011920929 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000019_obj0
[ [ 0.5285714268684387, 0.37460318207740784 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000019_obj1
[ [ 0.4214285612106323, 0.5492063760757446 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000020_obj0
[ [ 0.5205078125, 0.5124816298484802 ] ]
jellyfish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000020_obj1
[ [ 0.7724609375, 0.37444934248924255 ] ]
jellyfish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000021_obj0
[ [ 0.07083333283662796, 0.8281053900718689 ] ]
keyboard
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000021_obj1
[ [ 0.26750001311302185, 0.922208309173584 ] ]
table
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000022_obj0
[ [ 0.51171875, 0.947291374206543 ] ]
sidewalk
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000022_obj1
[ [ 0.818359375, 0.6939970850944519 ] ]
paper
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000022_obj2
[ [ 0.4228515625, 0.8316251635551453 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000022_obj3
[ [ 0.4072265625, 0.8301610350608826 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000023_obj0
[ [ 0.7587890625, 0.74609375 ] ]
penguin
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000023_obj1
[ [ 0.5732421875, 0.3697916567325592 ] ]
penguin
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000023_obj2
[ [ 0.3369140625, 0.46875 ] ]
penguin
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000023_obj3
[ [ 0.509765625, 0.1119791641831398 ] ]
sky
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000024_obj0
[ [ 0.025390625, 0.8697916865348816 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000024_obj1
[ [ 0.4755859375, 0.42578125 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000024_obj2
[ [ 0.7734375, 0.7278645634651184 ] ]
alga
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000024_obj3
[ [ 0.8251953125, 0.3111979067325592 ] ]
fish
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000025_obj0
[ [ 0.42158859968185425, 0.4177272617816925 ] ]
signboard
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000025_obj1
[ [ 0.10794297605752945, 0.5199999809265137 ] ]
tree
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000025_obj2
[ [ 0.08961303532123566, 0.5886363387107849 ] ]
car
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000025_obj3
[ [ 0.016972165554761887, 0.4572727382183075 ] ]
sky
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000026_obj0
[ [ 0.8554545640945435, 0.7672727108001709 ] ]
van
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000026_obj1
[ [ 0.5609090924263, 0.7987878918647766 ] ]
person
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000026_obj2
[ [ 0.5099999904632568, 0.9200000166893005 ] ]
sidewalk
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000026_obj3
[ [ 0.9954545497894287, 0 ] ]
arcade
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000027_obj0
[ [ 0.5363204479217529, 0.7118181586265564 ] ]
car
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
tr000027_obj1
[ [ 0.6028513312339783, 0.725454568862915 ] ]
column
Did anything change between the two images? Identify the object that changed and answer briefly (e.g., ‘the orange box’, ‘the brown vase on the bookshelf’).
Output the changed region as a binary mask.
Point to the changed object as [(x,y)] in normalized coordinates.
no
train
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
80