Picture for Dandan Xiao

Dandan Xiao

VA-AR: Learning Velocity-Aware Action Representations with Mixture of Window Attention

Add code
Mar 14, 2025
Viaarxiv icon