Hugging Face's logo

Datasets:
Haosonnn
/
OpenHumanVid-Talking

Tasks:
Text-to-Video
Video Classification
Modalities:
Audio
Image
Tabular
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
talking-head
camera-controlled
audio-conditioned
video-generation
monocular-slam
metric-depth
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
2
OpenHumanVid-Talking / annotations

Commit History

Fix active-speaker face/lip: per-frame speaker selection; drop match_seq==2; scalars = frame-0 aliases
9e06717
verified

Haosonnn commited on

Add derived training artifacts: alignnorm npz, per-frame lip sidecar, keep-list (8,884) (#3)
232dcd6

Haosonnn hao3on commited on

Switch annotations from MonST3R to MegaSaM 001-030 (metric, w2c, per-frame speaker face/lip)
61933d2
verified

Haosonnn commited on

Add monst3r pose + face/lip-bbox annotations (parts 001-040)
6bf8a60
verified

Haosonnn commited on