FreeCHROMA adapts a pretrained Wan-family video diffusion transformer using input-projection weights for foreground and mask conditioning, a background-reference convolution, and LoRA adapters.
Generative chroma keying from unaligned background references preserves foreground performance
FreeCHROMA adapts a pretrained video diffusion model to composite masked foregrounds into new environments using only a few unaligned background stills.
Big Tech
Jiayi Wu · Tianfu Wang · Rui Zhao · Florian Schiffers · Zhimin Chen · Cornelia Fermuller · +3 more
University of Maryland, College Park · Amazon · Arizona State University
Research Digest··2 min read
The authors introduce FreeCHROMA, a parameter-efficient video diffusion model that composites a masked foreground into a new environment from one to three unaligned background images.
Why this paper
From Amazon and 2 others
In one line
FreeCHROMA generates composite videos from unaligned background references while preserving foreground identity and motion.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors (2 noted)
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§