zeroscope_v1_320s / README.md
cerspense's picture
Update README.md
97c6804
metadata
tags:
  - Text-to-Video
license: cc-by-nc-4.0

Modelscope without the watermark, trained in 320x320 from the original weights, with no skipped frames for less flicker. See comparison here: https://www.youtube.com/watch?v=r4tOc30Zu0w Model was trained on a subset of the vimeo90k dataset + a selection of music videos