Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: ArrowInvalid
Message: JSON parse error: Invalid value. in row 0
Traceback: Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py", line 160, in _generate_tables
df = pandas_read_json(f)
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py", line 38, in pandas_read_json
return pd.read_json(path_or_buf, **kwargs)
File "/src/services/worker/.venv/lib/python3.9/site-packages/pandas/io/json/_json.py", line 791, in read_json
json_reader = JsonReader(
File "/src/services/worker/.venv/lib/python3.9/site-packages/pandas/io/json/_json.py", line 905, in __init__
self.data = self._preprocess_data(data)
File "/src/services/worker/.venv/lib/python3.9/site-packages/pandas/io/json/_json.py", line 917, in _preprocess_data
data = data.read()
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/utils/file_utils.py", line 827, in read_with_retries
out = read(*args, **kwargs)
File "/usr/local/lib/python3.9/codecs.py", line 322, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1855, in _prepare_split_single
for _, table in generator:
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 690, in wrapped
for item in generator(*args, **kwargs):
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py", line 163, in _generate_tables
raise e
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/packaged_modules/json/json.py", line 137, in _generate_tables
pa_table = paj.read_json(
File "pyarrow/_json.pyx", line 308, in pyarrow._json.read_json
File "pyarrow/error.pxi", line 154, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 91, in pyarrow.lib.check_status
pyarrow.lib.ArrowInvalid: JSON parse error: Invalid value. in row 0
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1431, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 992, in stream_convert_to_parquet
builder._prepare_split(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1742, in _prepare_split
for job_id, done, content in self._prepare_split_single(
File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1898, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
dataset
string | split
string | question_id
int64 | width
int64 | height
int64 | input_width
int64 | input_height
int64 | image
sequence | conversations
list |
|---|---|---|---|---|---|---|---|---|
textcap
|
train
| 1 | 1,024 | 1,024 | 616 | 616 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/c520b956136b72ce.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textcap/c520b956136b72ce.jpg###[343, 773, 676, 857].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat variety of wine is shown on the bottle label in the photo?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[206, 465, 406, 516]"
},
{
"from": "human",
"value": "<image>\nWhat variety of wine is shown on the bottle label in the photo?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "CHARBONO"
}
] |
gqa
|
train
| 2 | 800 | 539 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/4090.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/4090.jpg###[352, 205, 429, 267].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of furniture is the lamp on, a shelf or an end table?\" \nA. Shelf \nB. End table \nC. Coffee table \nD. Nightstand \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[332, 191, 405, 249]"
},
{
"from": "human",
"value": "<image>\nWhat kind of furniture is the lamp on, a shelf or an end table?\" \nA. Shelf \nB. End table \nC. Coffee table \nD. Nightstand \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
gqa
|
train
| 3 | 500 | 332 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2412718.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2412718.jpg###[0, 95, 500, 331].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the dog on?\"\nA. Couch \nB. Blanket \nC. Bed \nD. Floor \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 96, 504, 334]"
},
{
"from": "human",
"value": "<image>\nWhat is the dog on?\"\nA. Couch \nB. Blanket \nC. Bed \nD. Floor \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 4 | 500 | 483 | 504 | 476 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/150582765.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/150582765.jpg###[28, 17, 343, 310].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat distinctive feature does the woman's hair have?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[28, 16, 345, 305]"
},
{
"from": "human",
"value": "<image>\nWhat distinctive feature does the woman's hair have?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The woman has red curly hair."
}
] |
openimages
|
train
| 5 | 1,024 | 819 | 700 | 560 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/20e5c174b92b6949.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/20e5c174b92b6949.jpg###[353, 189, 684, 801].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho is on the bicycle?\" \nA. Girl \nB. Dog \nC. Boy \nD. Man \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[241, 129, 468, 547]"
},
{
"from": "human",
"value": "<image>\nWho is on the bicycle?\" \nA. Girl \nB. Dog \nC. Boy \nD. Man \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 6 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3441701164.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3441701164.jpg###[141, 265, 215, 311].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWithin the image, can you identify any logos or branding on the individual with the motorcycling suit?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[142, 267, 216, 313]"
},
{
"from": "human",
"value": "<image>\nWithin the image, can you identify any logos or branding on the individual with the motorcycling suit?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes, the individual is wearing a sponsor-covered motorcycling suit."
}
] |
openimages
|
train
| 7 | 1,024 | 768 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/0f13f1376e404c8b.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/0f13f1376e404c8b.jpg###[0, 570, 303, 767].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat material is the object that the wooden chair is made of?\"\nA. Metal \nB. Plastic \nC. Wood \nD. Glass \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 395, 215, 531]"
},
{
"from": "human",
"value": "<image>\nWhat material is the object that the wooden chair is made of?\"\nA. Metal \nB. Plastic \nC. Wood \nD. Glass \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 8 | 500 | 336 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2378361.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2378361.jpg###[329, 38, 461, 150].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is under the banana?\"\nA. Orange \nB. Grape \nC. Apple \nD. Kiwi \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[331, 38, 464, 150]"
},
{
"from": "human",
"value": "<image>\nWhat is under the banana?\"\nA. Orange \nB. Grape \nC. Apple \nD. Kiwi \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
docvqa
|
train
| 9 | 2,962 | 2,282 | 700 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/docvqa/mkmg0227_2.png",
"/your/path/to/Visual-CoT/cot_image_data_crop/docvqa/mkmg0227_2.png###[356, 411, 1552, 460].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat is the Emp. no. of Frances Ellen Brock ?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[84, 95, 366, 107]"
},
{
"from": "human",
"value": "<image>\nwhat is the Emp. no. of Frances Ellen Brock ?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "124232"
}
] |
textcap
|
train
| 10 | 478 | 1,024 | 420 | 924 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/26e3b184aa7fcba3.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textcap/26e3b184aa7fcba3.jpg###[246, 742, 451, 885].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the last name of the main character featured on the poster?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[216, 669, 396, 798]"
},
{
"from": "human",
"value": "<image>\nWhat is the last name of the main character featured on the poster?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Jones"
}
] |
flickr30k
|
train
| 11 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3162442331.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3162442331.jpg###[91, 279, 202, 333].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the texture of the material in the elephant's trunk look like?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[91, 281, 203, 336]"
},
{
"from": "human",
"value": "<image>\nWhat does the texture of the material in the elephant's trunk look like?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The material looks like the texture of grass."
}
] |
gqa
|
train
| 12 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2334443.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2334443.jpg###[75, 0, 500, 375].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the bear lying on?\" \nA. Grass \nB. Ground \nC. Snow \nD. Rock \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[75, 0, 504, 364]"
},
{
"from": "human",
"value": "<image>\nWhat is the bear lying on?\" \nA. Grass \nB. Ground \nC. Snow \nD. Rock \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
v7w
|
train
| 13 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/v7w/v7w_2401506.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/v7w/v7w_2401506.jpg###[119, 172, 341, 377].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat color is the man's shirt?\nA.It is purple.\nB.It is burgundy.\nC.It is neon.\nD.It is teal.\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[119, 166, 343, 365]"
},
{
"from": "human",
"value": "<image>\nWhat color is the man's shirt?\nA.It is purple.\nB.It is burgundy.\nC.It is neon.\nD.It is teal.\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
infographicsvqa
|
train
| 14 | 1,500 | 924 | 784 | 476 |
[
"/your/path/to/Visual-CoT/cot_image_data/infographicsvqa/38470.jpeg",
"/your/path/to/Visual-CoT/cot_image_data_crop/infographicsvqa/38470.jpeg###[444, 614, 699, 629].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat are Nextag, PriceGrabber and Shopzilla examples of?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[232, 316, 365, 324]"
},
{
"from": "human",
"value": "<image>\nWhat are Nextag, PriceGrabber and Shopzilla examples of?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Comparison Shopping Engines"
}
] |
flickr30k
|
train
| 15 | 500 | 341 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4958781334.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4958781334.jpg###[105, 40, 257, 160].jpg"
] |
[
{
"from": "human",
"value": "<image>\nCan you describe the headgear that the two individuals in the garden are wearing?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[105, 39, 259, 157]"
},
{
"from": "human",
"value": "<image>\nCan you describe the headgear that the two individuals in the garden are wearing?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "They are wearing traditional head garb."
}
] |
openimages
|
train
| 16 | 1,024 | 658 | 784 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/4c2f13532d8e1f0e.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/4c2f13532d8e1f0e.jpg###[439, 86, 1022, 656].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho wears the sunglasses?\" \nA. Boy \nB. Girl \nC. Man \nD. Dog \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[336, 66, 783, 503]"
},
{
"from": "human",
"value": "<image>\nWho wears the sunglasses?\" \nA. Boy \nB. Girl \nC. Man \nD. Dog \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
openimages
|
train
| 17 | 1,024 | 768 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/33c8aadd28a5482d.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/33c8aadd28a5482d.jpg###[515, 309, 711, 714].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho plays the musical keyboard?\" \nA. Man \nB. Woman \nC. Child \nD. Dog \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[366, 214, 505, 494]"
},
{
"from": "human",
"value": "<image>\nWho plays the musical keyboard?\" \nA. Man \nB. Woman \nC. Child \nD. Dog \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
gqa
|
train
| 18 | 500 | 334 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2385570.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2385570.jpg###[119, 226, 134, 275].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho holds the umbrella?\" \nA. Man \nB. Child \nC. Woman \nD. Dog \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[119, 227, 135, 276]"
},
{
"from": "human",
"value": "<image>\nWho holds the umbrella?\" \nA. Man \nB. Child \nC. Woman \nD. Dog \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 19 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2408683.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2408683.jpg###[198, 114, 332, 302].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhich kind of furniture is to the right of the chair?\" \nA. Table \nB. Couch \nC. Bookshelf \nD. Ottoman \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[199, 114, 334, 304]"
},
{
"from": "human",
"value": "<image>\nWhich kind of furniture is to the right of the chair?\" \nA. Table \nB. Couch \nC. Bookshelf \nD. Ottoman \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 20 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3834993910.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3834993910.jpg###[22, 102, 140, 314].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat attire has the man paired with his khaki shorts for his mini-golf game?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[22, 102, 141, 316]"
},
{
"from": "human",
"value": "<image>\nWhat attire has the man paired with his khaki shorts for his mini-golf game?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The man has paired a blue, large plaid print sweatshirt with his khaki shorts."
}
] |
flickr30k
|
train
| 21 | 328 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2938996469.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2938996469.jpg###[3, 54, 39, 133].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat activity is happening behind the man who is putting on gloves?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[3, 54, 39, 134]"
},
{
"from": "human",
"value": "<image>\nWhat activity is happening behind the man who is putting on gloves?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "People are setting up tables behind the man."
}
] |
gqa
|
train
| 22 | 500 | 323 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2383860.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2383860.jpg###[233, 0, 260, 50].jpg"
] |
[
{
"from": "human",
"value": "<image>\nThe person that is standing is holding what?\" \nA. A book \nB. A pole \nC. A bag \nD. A stick \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[234, 0, 262, 52]"
},
{
"from": "human",
"value": "<image>\nThe person that is standing is holding what?\" \nA. A book \nB. A pole \nC. A bag \nD. A stick \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
v7w
|
train
| 23 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/v7w/v7w_2325333.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/v7w/v7w_2325333.jpg###[97, 74, 103, 80].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is yellow?\nA.Lemon.\nB.The banana.\nC.Ball.\nD.The flag.\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[97, 74, 103, 80]"
},
{
"from": "human",
"value": "<image>\nWhat is yellow?\nA.Lemon.\nB.The banana.\nC.Ball.\nD.The flag.\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 24 | 500 | 334 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2647062476.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2647062476.jpg###[59, 72, 129, 313].jpg"
] |
[
{
"from": "human",
"value": "<image>\nIs there any visible barrier around the group of individuals in the field?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[59, 72, 130, 314]"
},
{
"from": "human",
"value": "<image>\nIs there any visible barrier around the group of individuals in the field?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes, there is a fence visible around the group of individuals in the field."
}
] |
gqa
|
train
| 25 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2358424.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2358424.jpg###[231, 90, 448, 234].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of animal walks on the tree?\"\nA. Squirrel \nB. Fish \nC. Bird \nD. Snake \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[232, 90, 451, 236]"
},
{
"from": "human",
"value": "<image>\nWhat kind of animal walks on the tree?\"\nA. Squirrel \nB. Fish \nC. Bird \nD. Snake \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 26 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2403409.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2403409.jpg###[0, 88, 239, 179].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of aircraft is to the left of the truck?\" \nA. Helicopters \nB. Gliders \nC. Airplanes \nD. Drones \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 88, 240, 180]"
},
{
"from": "human",
"value": "<image>\nWhat kind of aircraft is to the left of the truck?\" \nA. Helicopters \nB. Gliders \nC. Airplanes \nD. Drones \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
docvqa
|
train
| 27 | 1,664 | 2,166 | 532 | 700 |
[
"/your/path/to/Visual-CoT/cot_image_data/docvqa/gsfm0227_10.png",
"/your/path/to/Visual-CoT/cot_image_data_crop/docvqa/gsfm0227_10.png###[1308, 673, 1453, 731].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the Participation (Millions) for School Lunch?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[418, 217, 464, 236]"
},
{
"from": "human",
"value": "<image>\nWhat is the Participation (Millions) for School Lunch?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "25.8"
}
] |
gqa
|
train
| 28 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2404248.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2404248.jpg###[78, 239, 275, 453].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat animal is wearing the shirt?\"\nA. Lion \nB. Bear \nC. Tiger \nD. Kangaroo \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[78, 240, 277, 456]"
},
{
"from": "human",
"value": "<image>\nWhat animal is wearing the shirt?\"\nA. Lion \nB. Bear \nC. Tiger \nD. Kangaroo \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
docvqa
|
train
| 29 | 1,692 | 2,245 | 532 | 728 |
[
"/your/path/to/Visual-CoT/cot_image_data/docvqa/txvg0065_1.png",
"/your/path/to/Visual-CoT/cot_image_data_crop/docvqa/txvg0065_1.png###[1144, 283, 1253, 308].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the job number given?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[359, 91, 393, 99]"
},
{
"from": "human",
"value": "<image>\nWhat is the job number given?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "682"
}
] |
flickr30k
|
train
| 30 | 500 | 369 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/236095034.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/236095034.jpg###[212, 34, 310, 329].jpg"
] |
[
{
"from": "human",
"value": "<image>\nAre there any safety measures observed for the climber within the image?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[213, 33, 312, 324]"
},
{
"from": "human",
"value": "<image>\nAre there any safety measures observed for the climber within the image?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes, there are spotters positioned underneath the climber to ensure his safety."
}
] |
textvqa
|
train
| 31 | 1,024 | 765 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/58d02016eb332af7.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textvqa/58d02016eb332af7.jpg###[98, 522, 141, 544].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat is this man's name?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[70, 363, 100, 378]"
},
{
"from": "human",
"value": "<image>\nwhat is this man's name?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "noam chomsky"
}
] |
gqa
|
train
| 32 | 500 | 387 | 504 | 392 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2342727.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2342727.jpg###[163, 138, 444, 352].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat animal is standing in the dry grass?\" \nA. Giraffe \nB. Elephant \nC. Kangaroo \nD. Zebra \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[164, 139, 447, 356]"
},
{
"from": "human",
"value": "<image>\nWhat animal is standing in the dry grass?\" \nA. Giraffe \nB. Elephant \nC. Kangaroo \nD. Zebra \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
textcap
|
train
| 33 | 1,024 | 576 | 840 | 448 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/016ee5993838b383.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textcap/016ee5993838b383.jpg###[536, 300, 640, 402].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat number is visible on the side of the red airplane?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[439, 233, 525, 313]"
},
{
"from": "human",
"value": "<image>\nWhat number is visible on the side of the red airplane?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "27"
}
] |
flickr30k
|
train
| 34 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/1315116409.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/1315116409.jpg###[5, 4, 105, 495].jpg"
] |
[
{
"from": "human",
"value": "<image>\nCan you describe the movement of the roller coaster from what you observe in the image?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[5, 4, 105, 498]"
},
{
"from": "human",
"value": "<image>\nCan you describe the movement of the roller coaster from what you observe in the image?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The roller coaster is accelerating downwards vertically on a downward slide."
}
] |
dude
|
train
| 35 | 1,700 | 2,200 | 532 | 700 |
[
"/your/path/to/Visual-CoT/cot_image_data/dude/7254b13a6d3b6297f67dda0e43b7d99e_0.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/dude/7254b13a6d3b6297f67dda0e43b7d99e_0.jpg###[966, 2141, 1110, 2172].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhen is the approved for production date?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[302, 681, 347, 691]"
},
{
"from": "human",
"value": "<image>\nWhen is the approved for production date?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "12-28-59"
}
] |
flickr30k
|
train
| 36 | 375 | 500 | 364 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3349307529.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3349307529.jpg###[78, 429, 116, 478].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the color of the pants that the skier is wearing in the photo?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[75, 432, 112, 481]"
},
{
"from": "human",
"value": "<image>\nWhat is the color of the pants that the skier is wearing in the photo?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The skier is wearing bright yellow pants."
}
] |
gqa
|
train
| 37 | 375 | 500 | 364 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2325675.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2325675.jpg###[137, 67, 350, 484].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat animal is in the drawer?\" \nA. Dog \nB. Rabbit \nC. Cat \nD. Hamster \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[132, 67, 339, 487]"
},
{
"from": "human",
"value": "<image>\nWhat animal is in the drawer?\" \nA. Dog \nB. Rabbit \nC. Cat \nD. Hamster \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 38 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3769934267.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3769934267.jpg###[1, 92, 163, 314].jpg"
] |
[
{
"from": "human",
"value": "<image>\nHow does the woman who is shopping for produce appear to be dressed?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[1, 92, 164, 316]"
},
{
"from": "human",
"value": "<image>\nHow does the woman who is shopping for produce appear to be dressed?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The woman shopping for produce is wearing a gray sweater."
}
] |
flickr30k
|
train
| 39 | 500 | 374 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/7131011211.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/7131011211.jpg###[1, 335, 500, 374].jpg"
] |
[
{
"from": "human",
"value": "<image>\nApart from the umbrella, what else is the person doing while crossing the street?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[1, 326, 504, 364]"
},
{
"from": "human",
"value": "<image>\nApart from the umbrella, what else is the person doing while crossing the street?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The person is crossing the street at an intersection."
}
] |
openimages
|
train
| 40 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/574e17469f545fb6.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/574e17469f545fb6.jpg###[306, 156, 621, 638].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the person wearing the glasses doing?\"\nA. Sitting\nB. Running\nC. Standing\nD. Lying down \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[226, 115, 458, 470]"
},
{
"from": "human",
"value": "<image>\nWhat is the person wearing the glasses doing?\"\nA. Sitting\nB. Running\nC. Standing\nD. Lying down \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 41 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/713120.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/713120.jpg###[695, 268, 785, 387].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of appliance is to the left of the plate on the right?\" \nA. Toaster \nB. Blender \nC. Coffee maker \nD. Microwave \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[513, 197, 579, 285]"
},
{
"from": "human",
"value": "<image>\nWhat kind of appliance is to the left of the plate on the right?\" \nA. Toaster \nB. Blender \nC. Coffee maker \nD. Microwave \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
textcap
|
train
| 42 | 1,024 | 512 | 896 | 448 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/858089a1f53939c6.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textcap/858089a1f53939c6.jpg###[553, 444, 770, 506].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhich company's services are advertised in the image?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[484, 388, 674, 443]"
},
{
"from": "human",
"value": "<image>\nWhich company's services are advertised in the image?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Elevate Printing"
}
] |
gqa
|
train
| 43 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2356925.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2356925.jpg###[0, 216, 137, 232].jpg"
] |
[
{
"from": "human",
"value": "<image>\nThe grass is in front of what?\" \nA. Trees \nB. Buildings \nC. Cars \nD. Sidewalks \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 217, 138, 234]"
},
{
"from": "human",
"value": "<image>\nThe grass is in front of what?\" \nA. Trees \nB. Buildings \nC. Cars \nD. Sidewalks \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
gqa
|
train
| 44 | 500 | 331 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2397359.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2397359.jpg###[134, 48, 283, 330].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho is in front of the keyboard?\" \nA. Woman \nB. Child \nC. Man \nD. Robot \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[135, 48, 285, 334]"
},
{
"from": "human",
"value": "<image>\nWho is in front of the keyboard?\" \nA. Woman \nB. Child \nC. Man \nD. Robot \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 45 | 500 | 500 | 504 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2370743.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2370743.jpg###[244, 272, 409, 424].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat's the lock on?\" \nA. Door \nB. Cabinet \nC. Safe \nD. Drawer \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[245, 274, 412, 427]"
},
{
"from": "human",
"value": "<image>\nWhat's the lock on?\" \nA. Door \nB. Cabinet \nC. Safe \nD. Drawer \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
openimages
|
train
| 46 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/4081939714aab29f.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/4081939714aab29f.jpg###[457, 160, 1023, 406].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the standing man ride?\"\nA. Bicycle \nB. Car \nC. Horse \nD. Train \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[337, 118, 755, 299]"
},
{
"from": "human",
"value": "<image>\nWhat does the standing man ride?\"\nA. Bicycle \nB. Car \nC. Horse \nD. Train \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 47 | 500 | 499 | 504 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4859170265.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4859170265.jpg###[328, 136, 423, 455].jpg"
] |
[
{
"from": "human",
"value": "<image>\nBased on the attire of the blond individuals, can you infer what kind of event they might be participating in?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[330, 137, 426, 459]"
},
{
"from": "human",
"value": "<image>\nBased on the attire of the blond individuals, can you infer what kind of event they might be participating in?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "I cannot infer the exact nature of the event based solely on attire, but their coordinated activities with balloons and literature suggest they might be part of a promotional event or campaign."
}
] |
openimages
|
train
| 48 | 768 | 1,024 | 532 | 728 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/58845e63fd814d5a.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/58845e63fd814d5a.jpg###[0, 317, 376, 940].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the wine glass contain?\" \nA. Vodka \nB. Beer \nC. Whiskey \nD. Champagne \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 225, 260, 668]"
},
{
"from": "human",
"value": "<image>\nWhat does the wine glass contain?\" \nA. Vodka \nB. Beer \nC. Whiskey \nD. Champagne \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
gqa
|
train
| 49 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2319630.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2319630.jpg###[93, 197, 181, 256].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of food is to the left of the bowl?\"\nA. Salad \nB. Sandwich \nC. Soup \nD. Pizza \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[93, 191, 182, 248]"
},
{
"from": "human",
"value": "<image>\nWhat kind of food is to the left of the bowl?\"\nA. Salad \nB. Sandwich \nC. Soup \nD. Pizza \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
openimages
|
train
| 50 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/258c79ec2ccd120c.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/258c79ec2ccd120c.jpg###[446, 278, 818, 682].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the person wearing glasses doing?\" \nA. Sitting \nB. Lying down \nC. Standing \nD. Running \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[329, 205, 604, 503]"
},
{
"from": "human",
"value": "<image>\nWhat is the person wearing glasses doing?\" \nA. Sitting \nB. Lying down \nC. Standing \nD. Running \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
openimages
|
train
| 51 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/3c8c3e151f5f1588.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/3c8c3e151f5f1588.jpg###[172, 47, 853, 682].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the person wearing the necklace doing?\"\nA. Sitting\nB. Walking\nC. Standing\nD. Lying down \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[127, 34, 629, 503]"
},
{
"from": "human",
"value": "<image>\nWhat is the person wearing the necklace doing?\"\nA. Sitting\nB. Walking\nC. Standing\nD. Lying down \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 52 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/170704395.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/170704395.jpg###[92, 174, 235, 368].jpg"
] |
[
{
"from": "human",
"value": "<image>\nCan you describe the mood of the person by the train?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[92, 175, 237, 370]"
},
{
"from": "human",
"value": "<image>\nCan you describe the mood of the person by the train?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The person by the train is smiling, indicating a happy or joyful mood."
}
] |
openimages
|
train
| 53 | 1,024 | 730 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/3bf1efcbb5a669c3.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/3bf1efcbb5a669c3.jpg###[805, 217, 1023, 729].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the person wearing glasses doing?\"\nA. Running \nB. Sitting \nC. Standing \nD. Lying down \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[572, 158, 727, 531]"
},
{
"from": "human",
"value": "<image>\nWhat is the person wearing glasses doing?\"\nA. Running \nB. Sitting \nC. Standing \nD. Lying down \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
dude
|
train
| 54 | 1,700 | 2,200 | 532 | 700 |
[
"/your/path/to/Visual-CoT/cot_image_data/dude/f779827a95b43bfd3d4b7ec81b58d13b_0.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/dude/f779827a95b43bfd3d4b7ec81b58d13b_0.jpg###[680, 94, 847, 141].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat is the approved date?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[212, 29, 265, 44]"
},
{
"from": "human",
"value": "<image>\nwhat is the approved date?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "2002/10/30"
}
] |
v7w
|
train
| 55 | 1,024 | 679 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/v7w/v7w_61528.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/v7w/v7w_61528.jpg###[242, 81, 1021, 677].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat has two horns?\nA.The cow.\nB.The giraffe.\nC.The goat.\nD.The male sheep.\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[178, 60, 753, 502]"
},
{
"from": "human",
"value": "<image>\nWhat has two horns?\nA.The cow.\nB.The giraffe.\nC.The goat.\nD.The male sheep.\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 56 | 500 | 354 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2347570082.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2347570082.jpg###[257, 95, 325, 199].jpg"
] |
[
{
"from": "human",
"value": "<image>\nCan you identify the color of the clothing worn by the individuals who appear to be having a conversation?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[259, 97, 327, 204]"
},
{
"from": "human",
"value": "<image>\nCan you identify the color of the clothing worn by the individuals who appear to be having a conversation?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The individuals are wearing yellow clothing."
}
] |
flickr30k
|
train
| 57 | 375 | 500 | 364 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/5526795355.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/5526795355.jpg###[106, 145, 374, 488].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of clothing is the violinist wearing in the image?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[102, 146, 363, 491]"
},
{
"from": "human",
"value": "<image>\nWhat kind of clothing is the violinist wearing in the image?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The violinist is wearing a silky red outfit."
}
] |
openimages
|
train
| 58 | 683 | 1,024 | 504 | 756 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/1654080d689806e6.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/1654080d689806e6.jpg###[24, 380, 423, 758].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the standing girl interact with?\" \nA. Dog \nB. Cat \nC. Rabbit \nD. Bird \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[17, 281, 312, 559]"
},
{
"from": "human",
"value": "<image>\nWhat does the standing girl interact with?\" \nA. Dog \nB. Cat \nC. Rabbit \nD. Bird \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
openimages
|
train
| 59 | 1,024 | 768 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/326f2ac654af13f6.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/326f2ac654af13f6.jpg###[0, 388, 1023, 767].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhere is the bottle located?\"\nA. Chair \nB. Table \nC. Shelf \nD. Floor \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 269, 727, 531]"
},
{
"from": "human",
"value": "<image>\nWhere is the bottle located?\"\nA. Chair \nB. Table \nC. Shelf \nD. Floor \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
cub
|
train
| 60 | 230 | 178 | 224 | 168 |
[
"/your/path/to/Visual-CoT/cot_image_data/cub/007.Parakeet_Auklet/Parakeet_Auklet_0041_795933.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/cub/Parakeet_Auklet_0041_795933.jpg###[40, 22, 153, 173].jpg"
] |
[
{
"from": "human",
"value": "<image>\nDoes the bird in the picture have green nape and eyebrow head?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[38, 20, 149, 163]"
},
{
"from": "human",
"value": "<image>\nDoes the bird in the picture have green nape and eyebrow head?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "No"
}
] |
gqa
|
train
| 61 | 800 | 600 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/645.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/645.jpg###[471, 361, 752, 534].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho is commonly associated with wearing jackets?\" \nA. Animals \nB. People \nC. Inanimate objects \nD. Trees \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[428, 320, 684, 473]"
},
{
"from": "human",
"value": "<image>\nWho is commonly associated with wearing jackets?\" \nA. Animals \nB. People \nC. Inanimate objects \nD. Trees \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 62 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/1321723162.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/1321723162.jpg###[271, 45, 400, 326].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the apparent relationship between the older skater and the younger one?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[273, 43, 403, 316]"
},
{
"from": "human",
"value": "<image>\nWhat is the apparent relationship between the older skater and the younger one?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The older skater appears to be helping or instructing the younger one."
}
] |
infographicsvqa
|
train
| 63 | 1,527 | 825 | 840 | 448 |
[
"/your/path/to/Visual-CoT/cot_image_data/infographicsvqa/35919.jpeg",
"/your/path/to/Visual-CoT/cot_image_data_crop/infographicsvqa/35919.jpeg###[38, 395, 63, 404].jpg"
] |
[
{
"from": "human",
"value": "<image>\nBy which year the daily calories per person will be more than 3000?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[21, 214, 34, 219]"
},
{
"from": "human",
"value": "<image>\nBy which year the daily calories per person will be more than 3000?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "2050"
}
] |
gqa
|
train
| 64 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2349488.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2349488.jpg###[377, 0, 500, 86].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhich kind of furniture is to the right of the fence?\"\nA. Tables \nB. Sofas \nC. Chairs \nD. Benches \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[380, 0, 504, 86]"
},
{
"from": "human",
"value": "<image>\nWhich kind of furniture is to the right of the fence?\"\nA. Tables \nB. Sofas \nC. Chairs \nD. Benches \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 65 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4763276529.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4763276529.jpg###[289, 222, 334, 303].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhere is this encounter between the woman and the duck taking place?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[291, 224, 336, 305]"
},
{
"from": "human",
"value": "<image>\nWhere is this encounter between the woman and the duck taking place?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The encounter is happening in what looks like a parking lot."
}
] |
gqa
|
train
| 66 | 500 | 340 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2325486.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2325486.jpg###[44, 13, 478, 306].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat vehicle is not small?\"\nA. Bicycle \nB. Motorcycle \nC. Truck \nD. Scooter \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[44, 12, 481, 302]"
},
{
"from": "human",
"value": "<image>\nWhat vehicle is not small?\"\nA. Bicycle \nB. Motorcycle \nC. Truck \nD. Scooter \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
flickr30k
|
train
| 67 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3673484638.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3673484638.jpg###[47, 167, 170, 436].jpg"
] |
[
{
"from": "human",
"value": "<image>\nCan you describe what the children walking the runway are sporting in terms of attire?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[47, 168, 171, 439]"
},
{
"from": "human",
"value": "<image>\nCan you describe what the children walking the runway are sporting in terms of attire?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The children are showing off their fashion, which includes wearing denim."
}
] |
docvqa
|
train
| 68 | 1,700 | 2,200 | 532 | 700 |
[
"/your/path/to/Visual-CoT/cot_image_data/docvqa/kqyd0217_2.png",
"/your/path/to/Visual-CoT/cot_image_data_crop/docvqa/kqyd0217_2.png###[1253, 1523, 1419, 1554].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat is total- printing, solicitation and fulfillment ( 75,000 copies) amount?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[392, 484, 444, 494]"
},
{
"from": "human",
"value": "<image>\nwhat is total- printing, solicitation and fulfillment ( 75,000 copies) amount?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "$303,764.00"
}
] |
flickr30k
|
train
| 69 | 333 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/892340814.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/892340814.jpg###[134, 38, 249, 391].jpg"
] |
[
{
"from": "human",
"value": "<image>\nDoes the climber appear to be part of a group, or are they climbing alone?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[135, 38, 251, 394]"
},
{
"from": "human",
"value": "<image>\nDoes the climber appear to be part of a group, or are they climbing alone?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The climber is alone on the sheer cliff face; there are no other climbers visible in the vicinity."
}
] |
flickr30k
|
train
| 70 | 500 | 332 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4855382104.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4855382104.jpg###[5, 50, 494, 219].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat kind of setting is the man sleeping in? Is it quiet or busy?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[5, 50, 497, 221]"
},
{
"from": "human",
"value": "<image>\nWhat kind of setting is the man sleeping in? Is it quiet or busy?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The man is sleeping next to a busy street, suggesting the setting is not quiet."
}
] |
flickr30k
|
train
| 71 | 328 | 500 | 336 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/50162583.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/50162583.jpg###[104, 227, 237, 443].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat type of clothing is the woman wearing on her lower body?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[106, 228, 242, 446]"
},
{
"from": "human",
"value": "<image>\nWhat type of clothing is the woman wearing on her lower body?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The woman is wearing blue jeans."
}
] |
flickr30k
|
train
| 72 | 500 | 437 | 504 | 448 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/3362189985.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/3362189985.jpg###[119, 225, 290, 437].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the man in the red shirt carrying?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[119, 230, 292, 448]"
},
{
"from": "human",
"value": "<image>\nWhat is the man in the red shirt carrying?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The man is carrying a black shoulder bag."
}
] |
gqa
|
train
| 73 | 500 | 386 | 504 | 392 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2331153.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2331153.jpg###[0, 77, 500, 384].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhere are the shoes?\"\nA. On the shelf \nB. Under the bed \nC. Ground \nD. In the closet \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 78, 504, 389]"
},
{
"from": "human",
"value": "<image>\nWhere are the shoes?\"\nA. On the shelf \nB. Under the bed \nC. Ground \nD. In the closet \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
infographicsvqa
|
train
| 74 | 960 | 693 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/infographicsvqa/38564.jpeg",
"/your/path/to/Visual-CoT/cot_image_data_crop/infographicsvqa/38564.jpeg###[595, 603, 751, 623].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat percentage of women are in leadership roles in Fortune 500 companies, 23%, 4%, or 13.5%?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[451, 462, 569, 479]"
},
{
"from": "human",
"value": "<image>\nWhat percentage of women are in leadership roles in Fortune 500 companies, 23%, 4%, or 13.5%?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "4%"
}
] |
flickr30k
|
train
| 75 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/509912302.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/509912302.jpg###[179, 98, 342, 375].jpg"
] |
[
{
"from": "human",
"value": "<image>\nHow are the people in the image descending?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[180, 95, 344, 364]"
},
{
"from": "human",
"value": "<image>\nHow are the people in the image descending?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The people are descending on a movable staircase, an escalator."
}
] |
textvqa
|
train
| 76 | 1,024 | 668 | 784 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/3aa429d407212664.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textvqa/3aa429d407212664.jpg###[505, 190, 535, 223].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat number is on the watch?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[387, 143, 410, 168]"
},
{
"from": "human",
"value": "<image>\nwhat number is on the watch?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "12"
}
] |
cub
|
train
| 77 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/cub/109.American_Redstart/American_Redstart_0033_103783.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/cub/American_Redstart_0033_103783.jpg###[174, 107, 422, 253].jpg"
] |
[
{
"from": "human",
"value": "<image>\nDoes the bird in the picture have black crown and black eye?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[175, 107, 425, 255]"
},
{
"from": "human",
"value": "<image>\nDoes the bird in the picture have black crown and black eye?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes"
}
] |
flickr30k
|
train
| 78 | 351 | 500 | 364 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2588456052.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2588456052.jpg###[148, 220, 236, 309].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat color is the top the tennis player is wearing in the photo?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[153, 221, 244, 311]"
},
{
"from": "human",
"value": "<image>\nWhat color is the top the tennis player is wearing in the photo?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The tennis player is wearing a purple top."
}
] |
flickr30k
|
train
| 79 | 500 | 334 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/7489655004.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/7489655004.jpg###[122, 115, 192, 215].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat type of clothing is the female painter wearing in the photograph?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[122, 115, 193, 216]"
},
{
"from": "human",
"value": "<image>\nWhat type of clothing is the female painter wearing in the photograph?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The female painter is wearing a tan shirt and blue shorts."
}
] |
flickr30k
|
train
| 80 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4808226861.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4808226861.jpg###[250, 58, 333, 103].jpg"
] |
[
{
"from": "human",
"value": "<image>\nIs the man using the microphone at the concert wearing anything on his face?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[252, 58, 335, 103]"
},
{
"from": "human",
"value": "<image>\nIs the man using the microphone at the concert wearing anything on his face?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes, he is wearing sunglasses."
}
] |
flickr30k
|
train
| 81 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4631310439.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4631310439.jpg###[390, 1, 424, 50].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the common action performed by the group of individuals in the photograph?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[393, 1, 427, 50]"
},
{
"from": "human",
"value": "<image>\nWhat is the common action performed by the group of individuals in the photograph?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The group of individuals is staring upwards, possibly focused on something out of the frame."
}
] |
flickr30k
|
train
| 82 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2249034258.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2249034258.jpg###[250, 198, 264, 217].jpg"
] |
[
{
"from": "human",
"value": "<image>\nIs there anything distinctive about the way the woman is holding her mobile device?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[252, 199, 266, 218]"
},
{
"from": "human",
"value": "<image>\nIs there anything distinctive about the way the woman is holding her mobile device?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "Yes, the woman is holding the cellphone up to her ear, indicating she is actively using it."
}
] |
flickr30k
|
train
| 83 | 500 | 328 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/4641294847.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/4641294847.jpg###[248, 81, 326, 231].jpg"
] |
[
{
"from": "human",
"value": "<image>\nDoes the image give the impression that the man is indoors or outdoors?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[249, 82, 328, 236]"
},
{
"from": "human",
"value": "<image>\nDoes the image give the impression that the man is indoors or outdoors?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The man appears to be indoors, looking out through an above ground window."
}
] |
v7w
|
train
| 84 | 600 | 800 | 532 | 728 |
[
"/your/path/to/Visual-CoT/cot_image_data/v7w/v7w_713456.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/v7w/v7w_713456.jpg###[45, 287, 102, 333].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is everyone wearing?\nA.Uniforms.\nB.Jeans.\nC.Hoodies.\nD.Jackets and headgear.\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[39, 261, 90, 303]"
},
{
"from": "human",
"value": "<image>\nWhat is everyone wearing?\nA.Uniforms.\nB.Jeans.\nC.Hoodies.\nD.Jackets and headgear.\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "D"
}
] |
gqa
|
train
| 85 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2390243.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2390243.jpg###[109, 154, 147, 279].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho is wearing a hat?\" \nA. Girl \nB. Dog \nC. Boy \nD. Cat \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[109, 149, 148, 270]"
},
{
"from": "human",
"value": "<image>\nWho is wearing a hat?\" \nA. Girl \nB. Dog \nC. Boy \nD. Cat \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
infographicsvqa
|
train
| 86 | 1,500 | 2,765 | 448 | 840 |
[
"/your/path/to/Visual-CoT/cot_image_data/infographicsvqa/41638.jpeg",
"/your/path/to/Visual-CoT/cot_image_data_crop/infographicsvqa/41638.jpeg###[459, 2492, 854, 2520].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the Twitter handle name given?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[137, 757, 255, 765]"
},
{
"from": "human",
"value": "<image>\nWhat is the Twitter handle name given?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "AgencyFish"
}
] |
openimages
|
train
| 87 | 731 | 1,024 | 532 | 728 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/41fb78ac5b8bae0c.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/41fb78ac5b8bae0c.jpg###[0, 72, 731, 1024].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho wears the glasses?\" \nA. Woman \nB. Child \nC. Man \nD. Dog \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 51, 532, 728]"
},
{
"from": "human",
"value": "<image>\nWho wears the glasses?\" \nA. Woman \nB. Child \nC. Man \nD. Dog \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 88 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2404589.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2404589.jpg###[98, 15, 297, 333].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhere is the rug located?\"\nA. Living room \nB. Bathroom \nC. Kitchen \nD. Bedroom \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[98, 15, 299, 336]"
},
{
"from": "human",
"value": "<image>\nWhere is the rug located?\"\nA. Living room \nB. Bathroom \nC. Kitchen \nD. Bedroom \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
openimages
|
train
| 89 | 1,024 | 1,024 | 616 | 616 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/1aa9bb24a9261b4e.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/1aa9bb24a9261b4e.jpg###[183, 791, 305, 1015].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat material is the object that is at the wooden table made of?\"\nA. Plastic \nB. Metal \nC. Wood \nD. Glass \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[110, 476, 183, 610]"
},
{
"from": "human",
"value": "<image>\nWhat material is the object that is at the wooden table made of?\"\nA. Plastic \nB. Metal \nC. Wood \nD. Glass \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
openimages
|
train
| 90 | 1,024 | 683 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/5335f4eb1177cb33.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/5335f4eb1177cb33.jpg###[354, 231, 784, 520].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the standing man hold?\" \nA. Violin \nB. Trumpet \nC. Guitar \nD. Flute \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[261, 170, 578, 383]"
},
{
"from": "human",
"value": "<image>\nWhat does the standing man hold?\" \nA. Violin \nB. Trumpet \nC. Guitar \nD. Flute \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
docvqa
|
train
| 91 | 1,157 | 1,750 | 504 | 756 |
[
"/your/path/to/Visual-CoT/cot_image_data/docvqa/lpbm0227_10.png",
"/your/path/to/Visual-CoT/cot_image_data_crop/docvqa/lpbm0227_10.png###[116, 514, 940, 550].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat is the Official language of Congress?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[50, 222, 409, 237]"
},
{
"from": "human",
"value": "<image>\nWhat is the Official language of Congress?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "English"
}
] |
textvqa
|
train
| 92 | 1,024 | 505 | 896 | 420 |
[
"/your/path/to/Visual-CoT/cot_image_data/textvqa/8ff5868ad98f693c.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/textvqa/8ff5868ad98f693c.jpg###[819, 287, 986, 352].jpg"
] |
[
{
"from": "human",
"value": "<image>\nwhat kind of food does this store sell?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[717, 239, 863, 293]"
},
{
"from": "human",
"value": "<image>\nwhat kind of food does this store sell?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "cones, sundaes, shakes"
}
] |
flickr30k
|
train
| 93 | 500 | 500 | 504 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/1458395412.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/1458395412.jpg###[104, 128, 464, 334].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat color is the man's shirt who is sleeping in the chair?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[104, 129, 467, 336]"
},
{
"from": "human",
"value": "<image>\nWhat color is the man's shirt who is sleeping in the chair?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The man's shirt is gray."
}
] |
openimages
|
train
| 94 | 1,024 | 681 | 756 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/32cbeee2fb8ebcf9.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/32cbeee2fb8ebcf9.jpg###[0, 0, 1023, 680].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho wears the glasses?\" \nA. woman \nB. child \nC. man \nD. dog \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 0, 755, 503]"
},
{
"from": "human",
"value": "<image>\nWho wears the glasses?\" \nA. woman \nB. child \nC. man \nD. dog \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
gqa
|
train
| 95 | 500 | 375 | 504 | 364 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2404013.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2404013.jpg###[116, 63, 189, 264].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWho is kicking the boots?\" \nA. Boy \nB. Girl \nC. Dog \nD. Man \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[116, 61, 190, 256]"
},
{
"from": "human",
"value": "<image>\nWho is kicking the boots?\" \nA. Boy \nB. Girl \nC. Dog \nD. Man \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
flickr30k
|
train
| 96 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/10459869.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/10459869.jpg###[70, 117, 449, 333].jpg"
] |
[
{
"from": "human",
"value": "<image>\nAre the individuals in the image seated or standing while eating?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[70, 118, 452, 336]"
},
{
"from": "human",
"value": "<image>\nAre the individuals in the image seated or standing while eating?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The individuals are seated around a table as they eat."
}
] |
flickr30k
|
train
| 97 | 471 | 500 | 476 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/flickr30k/2227179257.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/flickr30k/2227179257.jpg###[60, 240, 160, 326].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the environment look like where the child is standing?\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[60, 241, 161, 328]"
},
{
"from": "human",
"value": "<image>\nWhat does the environment look like where the child is standing?\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "The child is amongst a snowy background, indicating it's a wintry setting outside."
}
] |
gqa
|
train
| 98 | 500 | 333 | 504 | 336 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2370992.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2370992.jpg###[84, 29, 427, 330].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhich device is on, the screen or the mobile phone?\" \nA. Tablet \nB. Desktop Computer \nC. Cell Phone \nD. Smartwatch \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[84, 29, 430, 332]"
},
{
"from": "human",
"value": "<image>\nWhich device is on, the screen or the mobile phone?\" \nA. Tablet \nB. Desktop Computer \nC. Cell Phone \nD. Smartwatch \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "C"
}
] |
openimages
|
train
| 99 | 1,024 | 746 | 728 | 532 |
[
"/your/path/to/Visual-CoT/cot_image_data/openimages/1de22fad10250523.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/openimages/1de22fad10250523.jpg###[568, 159, 692, 209].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat does the standing man wear?\"\nA. Hat \nB. Glasses \nC. Scarf \nD. Earrings \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[404, 113, 492, 149]"
},
{
"from": "human",
"value": "<image>\nWhat does the standing man wear?\"\nA. Hat \nB. Glasses \nC. Scarf \nD. Earrings \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
gqa
|
train
| 100 | 375 | 500 | 364 | 504 |
[
"/your/path/to/Visual-CoT/cot_image_data/gqa/2379534.jpg",
"/your/path/to/Visual-CoT/cot_image_data_crop/gqa/2379534.jpg###[1, 0, 150, 498].jpg"
] |
[
{
"from": "human",
"value": "<image>\nWhat vehicle is wet?\" \nA. Bus \nB. Train \nC. Bicycle \nD. Car \n\n Please only provide the bounding box coordinate of the region that can help you answer the question better. Following [x1,y1,x2,y2] format."
},
{
"from": "gpt",
"value": "[0, 0, 145, 501]"
},
{
"from": "human",
"value": "<image>\nWhat vehicle is wet?\" \nA. Bus \nB. Train \nC. Bicycle \nD. Car \n\n You can see the original image and the cropped image based on the bounding box you provided earlier. By referring to the original image and the image cropped from your previous bounding box, please try to give your final answer of the question."
},
{
"from": "gpt",
"value": "B"
}
] |
End of preview.
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
Data for Ground-R1 Training & VisCoT Benchmark: Visual-CoT.zip (We use a subset of: https://huggingface.co/datasets/deepcs233/Visual-CoT)
Data for LVLM benchmarks evaluation: https://huggingface.co/datasets/ZHZ2002/Ground_R1_General_Eval
Data for Grounding benchmarks evaluation: https://huggingface.co/datasets/ZHZ2002/RefCOCO_ALL_Eval
- Downloads last month
- 58