Open Datasets - a Norm Collection

Norm 's Collections

VAE

Image / Video Gen

Multimodal Language Model

Fundamental Research

Computer Vision

Open Datasets

updated 23 days ago

Thank you for sharing your dataset. I’ve fed them to my model, and they are benefit to it.

vikhyatk/lnqa

Viewer • Updated Aug 18 • 303k • 1.16k • 85

Note VQA
lmms-lab/LLaVA-OneVision-Data

Viewer • Updated Oct 22 • 3.72M • 10.1k • 150
naver-clova-ix/synthdog-en

Viewer • Updated Jan 31 • 66k • 6.32k • 17

Note OCR
Mutonix/Vript_Chinese

Viewer • Updated Oct 16 • 294k • 1.12k • 9

Note Caption, Video
ShareGPT4Video/ShareGPT4Video

Viewer • Updated Jul 8 • 40.2k • 3.07k • 185

Note Caption, Video
FreedomIntelligence/ALLaVA-4V

Viewer • Updated Sep 2 • 143k • 478 • 74

Note VQA, Caption
BAAI/DenseFusion-1M

Viewer • Updated Oct 17 • 1.18M • 901 • 29
tomg-group-umd/pixelprose

Viewer • Updated Jun 23 • 15.6M • 462 • 134
priyank-m/MJSynth_text_recognition

Viewer • Updated Jul 4, 2023 • 8.92M • 403 • 3

Note OCR
Rich Human Feedback for Text-to-Image Generation

Paper • 2312.10240 • Published Dec 15, 2023 • 19

Note Aesthetic Benchmark
HuggingFaceFV/finevideo

Viewer • Updated 15 days ago • 39.5k • 5.53k • 284
lmms-lab/LLaVA-Video-178K

Viewer • Updated Oct 11 • 1.63M • 15.1k • 97

Note Video Caption; Video QA Level-1: Every 10 seconds; Level-2: Every 30 seconds; Level-3: encapsulate the entire video.
lmms-lab/llava-critic-113k

Viewer • Updated Oct 5 • 113k • 767 • 25

Note DPO; Reflection
Salesforce/blip3-kale

Viewer • Updated Nov 14 • 235M • 4.87k • 33

Note Dense Caption
Koala-36M/Koala-36M-v1

Viewer • Updated Oct 12 • 36M • 381 • 25