Skip to content

UVA-Computer-Vision-Lab/FrameINO

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

4 Commits
Β 
Β 

Repository files navigation

Frame In-N-Out: Unbounded Controllable Image-to-Video Generation

Paper Website

We propose Frame In-N-Out, a controllable image-to-video generation framework where objects can enter or exit the scene along user-defined motion trajectories. Our method introduces a new dataset curation pattern recognition, evaluation protocol, and a motion-controllable, identity-preserving video Diffusion Transformer, to achieve Frame In and Frame Out in the cinematic domain.

Update πŸ”₯πŸ”₯πŸ”₯

  • Release the paper
  • Release the model weights (CogVideoX)
  • Gradio demo (with online)
  • Release the Training Code
  • Release the Processed Training Dataset

About

Frame In-N-Out: Unbounded Controllable Image-to-Video Generation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published