Datasets:

VCLab-PolyU
/

OmniCompass

Tasks:

Visual Question Answering

Video-Text-to-Text

Modalities:

Formats:

Languages:

Size:

ArXiv:

Tags:

video-understanding

temporal-grounding

spatial-grounding

Libraries:

License:

Dataset card Data Studio Files Files and versions

5.29 GB

Ctrl+K

Ctrl+K

1 contributor

History: 9 commits

VCLab-PolyU's picture

Add first two fully staged media archives (incremental)

d1cb84e verified 18 days ago

assets
Add dataset card and paper figures 19 days ago
media
Add first two fully staged media archives (incremental) 18 days ago
skills
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago
.gitattributes

2.5 kB
initial commit 19 days ago
README.md

6.68 kB
Describe dataset files with an annotated directory tree 19 days ago
archives.json

7.87 kB
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago
media_manifest.jsonl

670 kB
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago
questions.jsonl

1.67 MB
Add frozen annotations and media indexes without media archives 19 days ago
release.json

1.19 kB
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago
schema.json

9.67 kB
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago
taxonomy.json

5.96 kB
Add frozen annotations and media indexes without media archives 19 days ago
videos.jsonl

263 kB
Publish embedded-audio metadata and MP4 evaluation guidance 19 days ago