2 papers
cs.CV2026
Real2SAM2Real: Generative 3D Caches as Complementary Context for Video Diffusion
Jiayi Wu, Haoming Cai, Cornelia Fermuller +2
While Video Diffusion Models (VDMs) excel at synthesizing high-fidelity videos, enabling precise camera and scene control remains challenging. Existing methods predominantly rely o…
cs.CV2025
Parametric Shadow Control for Portrait Generation in Text-to-Image Diffusion Models
Haoming Cai, Tsung-Wei Huang, Shiv Gehlot +4
Text-to-image diffusion models excel at generating diverse portraits, but lack intuitive shadow control. Existing editing approaches, as post-processing, struggle to offer effectiv…