Skip to content

Latest commit

 

History

History
11 lines (9 loc) · 528 Bytes

README.md

File metadata and controls

11 lines (9 loc) · 528 Bytes

TTV

This repository enhances modelscope-damo-text-to-video-synthesis by adding VTV functionality, while also optimizing memory usage. By decoding frames sequentially instead of in batches, it enables larger image resolutions and longer video durations.

VTV

  1. A cow running a marathon
  2. A cow running a marathon in hell
  3. A dog running a marathon