CAKE: Real-time Action Detection via Motion Distillation and Background-aware Contrastive Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911543499685888 |
|---|---|
| author | Hoang, Hieu Tran, Dung Trung Nguyen, Hong Nguyen, Nam-Phong |
| author_facet | Hoang, Hieu Tran, Dung Trung Nguyen, Hong Nguyen, Nam-Phong |
| contents | 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. The framework also integrates a Floating Contrastive Learning strategy to distinguish informative motion dynamics from temporal background. Various experiments conducted on the TVSeries, THUMOS'14, Kinetics-400 datasets show effectiveness of our model. CAKE achieves a standout mAP compared with SOTA while using the same backbone. Our model operates at over 72 FPS on a single CPU, making it highly suitable for resource-constrained systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_23988 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CAKE: Real-time Action Detection via Motion Distillation and Background-aware Contrastive Learning Hoang, Hieu Tran, Dung Trung Nguyen, Hong Nguyen, Nam-Phong Computer Vision and Pattern Recognition 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. The framework also integrates a Floating Contrastive Learning strategy to distinguish informative motion dynamics from temporal background. Various experiments conducted on the TVSeries, THUMOS'14, Kinetics-400 datasets show effectiveness of our model. CAKE achieves a standout mAP compared with SOTA while using the same backbone. Our model operates at over 72 FPS on a single CPU, making it highly suitable for resource-constrained systems. |
| title | CAKE: Real-time Action Detection via Motion Distillation and Background-aware Contrastive Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.23988 |