PixFoundation: Are We Heading in the Right Direction with Pixel-level Vision Foundation Models?
Paper
•
2502.04192
•
Published
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 18961 new columns ({'60].6', 'date_captured:""}.157', '148.19', '113.17', '86.67', '93.8', 'iscrowd:0.316', '54.38', 'width:224.144', '204.10', 'id:297', '83.65', 'bbox:[64.1', 'area:1140', '144]].2', '{"segmentation":[[109.1', '139.81', 'license:0.75', 'file_name:"242.jpg"', 'date_captured:""}.83', 'iscrowd:0.321', '223.86', '159.49', '149.18', 'image_id:130.1', '144.25', 'iscrowd:0.54', '105].2', '83.31', 'license:0.139', 'iscrowd:0.234', '122.7', '196.12', '148.17', '133.62', '9.1', '167.12', '32.14', '11.11', '31.10', '125.23', '24.26', '87.79', '107.43', '82.60', '110.52', 'bbox:[77.1', '104.76', '67.11', '30.36', 'image_id:228.3', '51.7', '127.36', '141.55', '69.49', '16.14', '90.41', 'file_name:"10.jpg"', '104.4', '174.46', 'width:224.85', '{"id":282', 'date_captured:""}.117', 'license:0.289', 'license:0.95', 'width:224.242', '97.3', '144.45', '129.31', 'file_name:"65.jpg"', '166]]', '62.49', '{"id":31', '109.62', '103.39', '93.59', '222.65', '127.28', '92.21', 'image_id:219', '117.8', '134.16', '150.26', 'license:0.263', '125.51', '41.25', '118]]', 'image_id:236.1', '10.8', '23.20', '{"id":130', 'category_id:null}.38', '125.6', '73.13', '167.80', '131]', '107.5', '92]]', '13.3', '96.41', '114.39', '141.3', '36.23', '64.58', 'category_id:null}.376', '101.76', '132.2', '166.42', 'height:224.182', 'bbox:[56.3', 'date_captured:""}.113', '66.65', '0]]', '19.23', 'height:224.55', '123.31', 'image_id:49', '139.45', '213.26', '167.11', 'width:224.68', '127.87', '60.38', 'image_id:13', '131.42',
...
:""}.1', '148.12', '88.35', '3.9', '184.8', '29.36', '131].1', '32.19', '201.31', 'license:0.236', '153.27', '21.37', '67.9', '182.40', '211.12', '158.30', 'date_captured:""}.182', '203.13', '151.60', 'id:51', 'iscrowd:0.91', '51.6', '107.8', 'date_captured:""}]', '96.21', '95.42', '118.6', '192.34', 'date_captured:""}.22', 'height:224.3', '40.27', '100.9', '108.26', '127.56', '113.30', '128.100', 'bbox:[152.3', '33.16', '37].1', 'id:302', '81.24', '199.46', '14.19', '223.31', '105.59', 'category_id:null}.203', '183.12', '118.29', 'bbox:[207', '87.55', '214.31', '23.3', '85.34', '1.62', 'height:224.247', '119.17', '184.29', 'image_id:209', '222.76', '176.33', '129.11', '40.33', '114.49', '84.10', 'bbox:[9', '126.43', 'license:0.228', '{"id":103', '63.17', '159.41', 'file_name:"89.jpg"', '124.76', '104.36', '194.24', 'height:224.1', 'category_id:null}.28', '80.62', '164.7', '171.9', '15].1', 'category_id:null}.237', '68.74', '58.4', '1]].9', '68.66', '44.44', '95.67', 'date_captured:""}.155', '176.13', '185.37', '183.19', '171.47', '199.48', 'iscrowd:0.68', '{"segmentation":[[108.1', '123.19', '204.37', '80.13', '192.12', '53.23', '127.34', '196.33', '106.20', 'license:0.179', '130.11', '165.14', '148.41', '220.28', '220]]', '{"segmentation":[[94.1', '142.4', '96]]', '161.16', '202]].2', 'width:224.139', '125.49', '72.25', '{"segmentation":[[161', '81].3', '196.31', '198.30', 'id:328', '83.46', 'iscrowd:0.64', '159.47', '182.2', '190.35', 'area:170', 'width:224.274', '182.43'}) and 2 missing columns ({' Object', 'Index'}).
This happened while the csv dataset builder was generating data using
hf://datasets/IVUlab/pixmmvp/Segmentations.json (at revision 2d2a87856e057b3fe0b6daba73b8c66800b69a50), [/tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/Objects.csv (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/Objects.csv), /tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/Segmentations.json (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/Segmentations.json), /tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/visual_patterns.csv (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/visual_patterns.csv)]
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
writer.write_table(table)
File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 674, in write_table
pa_table = table_cast(pa_table, self._schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
{"info":{"year":2024: null
version:"1.0": null
description:"VIA project exported to COCO format using VGG Image Annotator (http://www.robots.ox.ac. (... 23 chars omitted): null
contributor:"": null
url:"http://www.robots.ox.ac.uk/~vgg/software/via/": null
date_created:"Sat Dec 07 2024 09:04:59 GMT-0500 (Eastern Standard Time)"}: null
images:[{"id":1: null
width:224: null
height:224: null
file_name:"1.jpg": null
license:0: null
date_captured:""}: null
{"id":2: null
width:224.1: null
height:224.1: null
file_name:"2.jpg": null
license:0.1: null
date_captured:""}.1: null
{"id":3: null
width:224.2: null
height:224.2: null
file_name:"3.jpg": null
license:0.2: null
date_captured:""}.2: null
{"id":4: null
width:224.3: null
height:224.3: null
file_name:"4.jpg": null
license:0.3: null
date_captured:""}.3: null
{"id":5: null
width:224.4: null
height:224.4: null
file_name:"5.jpg": null
license:0.4: null
date_captured:""}.4: null
{"id":6: null
width:224.5: null
height:224.5: null
file_name:"6.jpg": null
license:0.5: null
date_captured:""}.5: null
{"id":7: null
width:224.6: null
height:224.6: null
file_name:"7.jpg": null
license:0.6: null
date_captured:""}.6: null
{"id":8: null
width:224.7: null
height:224.7: null
file_name:"8.jpg": null
license:0.7: null
date_captured:""}.7: null
{"id":9: null
width:224.8: null
height:224.8: null
file_name:"9.jpg": null
license:0.8: null
date_captured:""}.8: null
{"id":10: null
width:224.9: null
height:224.9: null
file_name:"10.jpg": null
license:0.9: nul
...
29: null
66.73: null
222.87: null
138.93: null
214.31: null
143.82: null
222.88: null
158.64: null
220.30: null
173.67: null
220.31: null
188.42: null
216.37: null
197.35: null
211.33: null
207.33: null
201.42: null
211.34: null
188.43: null
216.38: null
173.68: null
216.39: null
161.63: null
210.44: null
153.58: null
200.53: null
146.64: null
214.32: null
128.101: null
222.89: null
105.86: null
222.90: null
75.69: null
217.27: null
61.52: null
200.54: null
54.44: null
184.50: null
53.67: null
173.69: null
45.43: null
187.61: null
34.55: null
192.46: null
16.22: null
192.47: null
1.107: null
185.48: null
2.46: null
126.96: null
6.21: null
115.77: null
19.23: null
111.84: null
25.27: null
108.76: null
24.30: null
99.52: null
19.24: null
80.86: null
19.25: null
68.79: null
28.31: null
54.45: null
39.36: null
45.44: null
59.65: null
43.48: null
70.81: null
44]].1: null
area:42370: null
bbox:[1.36: null
32.62: null
223.183: null
190].1: null
iscrowd:0.383: null
id:384: null
image_id:298: null
category_id:null}]: null
licenses:[{"id":0: null
name:"Unknown License": null
url:""}]: null
categories:[{"supercategory":"type": null
id:null: null
name:"Bird"}: null
{"supercategory":"type": null
id:null.1: null
name:"Human"}: null
{"supercategory":"type".1: null
id:null.2: null
name:"Cup (object)"}: null
{"supercategory":"type".2: null
id:null.3: null
name:"Unknown (object)"}]}: null
-- schema metadata --
pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 2142241
to
{'Index': Value('int64'), ' Object': Value('string')}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 884, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 947, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1736, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1889, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 18961 new columns ({'60].6', 'date_captured:""}.157', '148.19', '113.17', '86.67', '93.8', 'iscrowd:0.316', '54.38', 'width:224.144', '204.10', 'id:297', '83.65', 'bbox:[64.1', 'area:1140', '144]].2', '{"segmentation":[[109.1', '139.81', 'license:0.75', 'file_name:"242.jpg"', 'date_captured:""}.83', 'iscrowd:0.321', '223.86', '159.49', '149.18', 'image_id:130.1', '144.25', 'iscrowd:0.54', '105].2', '83.31', 'license:0.139', 'iscrowd:0.234', '122.7', '196.12', '148.17', '133.62', '9.1', '167.12', '32.14', '11.11', '31.10', '125.23', '24.26', '87.79', '107.43', '82.60', '110.52', 'bbox:[77.1', '104.76', '67.11', '30.36', 'image_id:228.3', '51.7', '127.36', '141.55', '69.49', '16.14', '90.41', 'file_name:"10.jpg"', '104.4', '174.46', 'width:224.85', '{"id":282', 'date_captured:""}.117', 'license:0.289', 'license:0.95', 'width:224.242', '97.3', '144.45', '129.31', 'file_name:"65.jpg"', '166]]', '62.49', '{"id":31', '109.62', '103.39', '93.59', '222.65', '127.28', '92.21', 'image_id:219', '117.8', '134.16', '150.26', 'license:0.263', '125.51', '41.25', '118]]', 'image_id:236.1', '10.8', '23.20', '{"id":130', 'category_id:null}.38', '125.6', '73.13', '167.80', '131]', '107.5', '92]]', '13.3', '96.41', '114.39', '141.3', '36.23', '64.58', 'category_id:null}.376', '101.76', '132.2', '166.42', 'height:224.182', 'bbox:[56.3', 'date_captured:""}.113', '66.65', '0]]', '19.23', 'height:224.55', '123.31', 'image_id:49', '139.45', '213.26', '167.11', 'width:224.68', '127.87', '60.38', 'image_id:13', '131.42',
...
:""}.1', '148.12', '88.35', '3.9', '184.8', '29.36', '131].1', '32.19', '201.31', 'license:0.236', '153.27', '21.37', '67.9', '182.40', '211.12', '158.30', 'date_captured:""}.182', '203.13', '151.60', 'id:51', 'iscrowd:0.91', '51.6', '107.8', 'date_captured:""}]', '96.21', '95.42', '118.6', '192.34', 'date_captured:""}.22', 'height:224.3', '40.27', '100.9', '108.26', '127.56', '113.30', '128.100', 'bbox:[152.3', '33.16', '37].1', 'id:302', '81.24', '199.46', '14.19', '223.31', '105.59', 'category_id:null}.203', '183.12', '118.29', 'bbox:[207', '87.55', '214.31', '23.3', '85.34', '1.62', 'height:224.247', '119.17', '184.29', 'image_id:209', '222.76', '176.33', '129.11', '40.33', '114.49', '84.10', 'bbox:[9', '126.43', 'license:0.228', '{"id":103', '63.17', '159.41', 'file_name:"89.jpg"', '124.76', '104.36', '194.24', 'height:224.1', 'category_id:null}.28', '80.62', '164.7', '171.9', '15].1', 'category_id:null}.237', '68.74', '58.4', '1]].9', '68.66', '44.44', '95.67', 'date_captured:""}.155', '176.13', '185.37', '183.19', '171.47', '199.48', 'iscrowd:0.68', '{"segmentation":[[108.1', '123.19', '204.37', '80.13', '192.12', '53.23', '127.34', '196.33', '106.20', 'license:0.179', '130.11', '165.14', '148.41', '220.28', '220]]', '{"segmentation":[[94.1', '142.4', '96]]', '161.16', '202]].2', 'width:224.139', '125.49', '72.25', '{"segmentation":[[161', '81].3', '196.31', '198.30', 'id:328', '83.46', 'iscrowd:0.64', '159.47', '182.2', '190.35', 'area:170', 'width:224.274', '182.43'}) and 2 missing columns ({' Object', 'Index'}).
This happened while the csv dataset builder was generating data using
hf://datasets/IVUlab/pixmmvp/Segmentations.json (at revision 2d2a87856e057b3fe0b6daba73b8c66800b69a50), [/tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/Objects.csv (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/Objects.csv), /tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/Segmentations.json (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/Segmentations.json), /tmp/hf-datasets-cache/medium/datasets/66249055984868-config-parquet-and-info-IVUlab-pixmmvp-ed4e745b/hub/datasets--IVUlab--pixmmvp/snapshots/2d2a87856e057b3fe0b6daba73b8c66800b69a50/visual_patterns.csv (origin=hf://datasets/IVUlab/pixmmvp@2d2a87856e057b3fe0b6daba73b8c66800b69a50/visual_patterns.csv)]
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Index
int64 | Object
string |
|---|---|
1
|
the butterfly's wings
|
2
|
the butterfly's wings
|
3
|
the flame of the match
|
4
|
the flame of the match
|
5
|
the dog's face
|
6
|
the dog's face
|
7
|
the school bus
|
8
|
the school bus
|
9
|
the key ""Z""
|
10
|
the key ""Z""
|
11
|
the peacock's tail
|
12
|
the peacock's tail
|
13
|
an ear of corn
|
14
|
an ear of corn
|
15
|
a shadow on the flower
|
16
|
a shadow on the flower
|
17
|
the front of the school bus
|
18
|
the front of the school bus
|
19
|
the dorsal fin of the animal
|
20
|
the dorsal fin of the animal
|
21
|
the duck's entire beak
|
22
|
the duck's entire beak
|
23
|
None
|
24
|
None
|
25
|
the chicken's body
|
26
|
the chicken's body
|
27
|
the peacock's head
|
28
|
the peacock's head
|
29
|
the window on the school bus
|
30
|
the window on the school bus
|
31
|
the flag
|
32
|
the flag
|
33
|
the butterfly's abdomen
|
34
|
the butterfly's abdomen
|
35
|
the vegetables with spikes
|
36
|
the vegetables with spikes
|
37
|
the flowers in the background
|
38
|
the flowers in the background
|
39
|
the duck
|
40
|
the duck
|
41
|
the spider's legs
|
42
|
the spider's legs
|
43
|
the wheels of the school bus
|
44
|
the wheels of the school bus
|
45
|
the shark's belly
|
46
|
the shark's belly
|
47
|
the spots on the animal
|
48
|
the spots on the animal
|
49
|
the arrow keys
|
50
|
the arrow keys
|
51
|
the elephant's trunk
|
52
|
the elephant's trunk
|
53
|
the pills
|
54
|
the pills
|
55
|
the stems of bananas
|
56
|
the stems of bananas
|
57
|
the crocodile
|
58
|
the crocodile
|
59
|
the lock
|
60
|
the lock
|
61
|
the letter ""J""
|
62
|
the letter ""J""
|
63
|
one daisy that is under the shadow of a taller daisy
|
64
|
one daisy that is under the shadow of a taller daisy
|
65
|
the letter D
|
66
|
the letter D
|
67
|
the clouds
|
68
|
the clouds
|
69
|
the snake's tongue
|
70
|
the snake's tongue
|
71
|
the lock
|
72
|
the lock
|
73
|
the ground
|
74
|
the ground
|
75
|
the flower
|
76
|
the flower
|
77
|
the caudal fin of the shark
|
78
|
the caudal fin of the shark
|
79
|
the snake's head
|
80
|
the snake's head
|
81
|
a hammerhead shark
|
82
|
a hammerhead shark
|
83
|
the door of the truck cab
|
84
|
the door of the truck cab
|
85
|
the ears of the dog
|
86
|
the ears of the dog
|
87
|
a hand using the mouse
|
88
|
a hand using the mouse
|
89
|
people
|
90
|
people
|
91
|
words on the vehicle's lightbar
|
92
|
words on the vehicle's lightbar
|
93
|
accessory on the wrists
|
94
|
accessory on the wrists
|
95
|
the spider web
|
96
|
the spider web
|
97
|
the keyboard
|
98
|
the keyboard
|
99
|
the elephant's tusks
|
100
|
the elephant's tusks
|
End of preview.
YAML Metadata
Warning:
The task_categories "image-referring-segmentation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
PixMMVP Benchmark
Project Page | Paper | GitHub
The PixMMVP dataset augments the MMVP benchmark with referring expressions and corresponding segmentation masks for the objects of interest in their respective questions within the original VQA task.
The goal of this benchmark is to evaluate the pixel-level visual grounding and visual question answering capabilities of recent pixel-level MLLMs (e.g., OMG-Llava, Llava-G, GLAMM, and LISA).
Acknowledgements
I acknowledge the use of MMVP dataset's images and questions/choices part of building this dataset, the original MMVP.
Citation
Please cite the following work if you find the dataset useful:
@article{siam2025pixfoundation,
title={PixFoundation: Are We Heading in the Right Direction with Pixel-level Vision Foundation Models?},
author={Siam, Mennatullah},
journal={arXiv preprint arXiv:2502.04192},
year={2025}
}
- Downloads last month
- 24