datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Youtube-Common-First-600-Parquetpexel-0808-complete-final-testGithub Page: https://github.com/UmiMarch/OpenVideo
license: cc-by-4.0
task_categories:
- video-text-to-text
size_categories:
- 100K<n<1M
open_video_dataOpenVideo-Scene-Reasoning
OpenVideo-Scene-Reasoning
OpenVideo-Scene-Reasoning is a video understanding dataset containing 2,894 short video clips, where each sample consists of a 10-second video, five uniformly sampled frames, and a dense scene-level response describing the complete temporal sequence.
Rather than generating captions for individual frames independently, the responses are synthesized by jointly reasoning over the sampled frames to capture temporal progression, object interactions, actions… See the full description on the dataset page: https://huggingface.co/datasets/prithivMLmods/OpenVideo-Scene-Reasoning.prompttest-18videosYouTube-Commons-5G-RawYoutube-Common-First-600open-r1-video-4kVideo2BEV-Open
Video2BEV-Open
Yet another distribution of the dataset from the paper: 📜 Video2BEV
Package and Data Structure
Package
The dataset is archived using Zstandard compression, e.g., train-shard00.tar.zst or test-shard01.tar.zst.
Extract with the command
tar xaf train-shard00.tar.zst
Extract all
for file in {train,test}-shard*.tar.zst; do
tar xaf "$file"
done
Data Structure
A dataset entry looks like:
# /train
0000
├── drone
│ ├──… See the full description on the dataset page: https://huggingface.co/datasets/NinZeige/Video2BEV-Open.open-video-prompts
open-video-prompts · v0.0.1
Curated prompt recipes for OpenVideo / MiniMax H3.
Software card: https://huggingface.co/fei567/open-video (transfer target: open-video-ai/open-video)
GitHub: https://github.com/open-video-ai/open-video
Site: https://open-video.ai
Files under prompts/*.txt.
open_video_helmet
Open Video Molina
This is a growing open-source repository for multiview video captured across
different places. More locations and recording sequences may be added over time.
The current release contains GoPro recordings from three locations: forest,
predaia, and molina. Each location has six camera views (camera_00 through
camera_05). The GoPros are mounted together on a fixed multicamera rig, keeping
their relative arrangement approximately consistent during each capture.
The… See the full description on the dataset page: https://huggingface.co/datasets/sebothetramp/open_video_helmet.flow_open_video