1 paper · 2 filters
Chenyangguang Zhang, Botao Ye, Boqi Chen +4
Controllable video generation for complex hand-object interactions is a critical step toward building visual world models. However, existing methods often struggle to achieve fine-…