
DatasetRank #427 of 836 176 likes
pixelprose
by tomg-group-umd (organization)
From Pixels to Prose: A Large Dataset of Dense Image Captions [ arXiv paper ] | [ 🌮 image tars ] PixelProse is a comprehensive dataset of over 16M (million) synthetically generated captions, leveraging cutting-edge vision-language models (Gemini 1.0 Pro Vision) for detailed… (excerpt from the dataset card on Hugging Face)
Dataset facts
- Hub id
- tomg-group-umd/pixelprose
- Likes
- 176
- Downloads (last 30 days)
- 1,012
- Tasks
- Image to text, Text to image, Visual question answering
- Modality
- image, tabular, text
- Size
- 10M to 100M examples
- License
- CC BY 4.0 (cc-by-4.0)
- Access
- Open download
- Author
- tomg-group-umd · Organization
- Languages
- en
- Created
- Jun 14, 2024
- Last modified
- Dec 13, 2025