AI RESEARCH
CAKE: Real-time Action Detection via Motion Distillation and Background-aware Contrastive Learning
arXiv CS.CV
•
ArXi:2603.23988v1 Announce Type: new Online Action Detection (OAD) systems face two primary challenges: high computational cost and insufficient modeling of discriminative temporal dynamics against background motion. Adding optical flow could provides strong motion cues but it incurs significant computational overhead. We propose CAKE, a OAD Flow-based distillation framework to transfer motion knowledge into RGB models. We propose Dynamic Motion Adapter (DMA) to suppress static background noise and emphasize pixel changes, effectively approximating optical flow without explicit computation.