Fetching the paper…

Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model · Around