Prediction-Feedback DETR for Temporal Action Detection
This work addresses a specific bottleneck in transformer-based models for temporal action detection, which is incremental but improves performance for video analysis applications.
The paper tackles the attention collapse problem in cross-attention within DETR-based temporal action detection methods, proposing Prediction-Feedback DETR (Pred-DETR) to restore collapse and align attention with predictions, achieving state-of-the-art performance among DETR-based methods on benchmarks like THUMOS14, ActivityNet-v1.3, HACS, and FineAction.
Temporal Action Detection (TAD) is fundamental yet challenging for real-world video applications. Leveraging the unique benefits of transformers, various DETR-based approaches have been adopted in TAD. However, it has recently been identified that the attention collapse in self-attention causes the performance degradation of DETR for TAD. Building upon previous research, this paper newly addresses the attention collapse problem in cross-attention within DETR-based TAD methods. Moreover, our findings reveal that cross-attention exhibits patterns distinct from predictions, indicating a short-cut phenomenon. To resolve this, we propose a new framework, Prediction-Feedback DETR (Pred-DETR), which utilizes predictions to restore the collapse and align the cross- and self-attention with predictions. Specifically, we devise novel prediction-feedback objectives using guidance from the relations of the predictions. As a result, Pred-DETR significantly alleviates the collapse and achieves state-of-the-art performance among DETR-based methods on various challenging benchmarks including THUMOS14, ActivityNet-v1.3, HACS, and FineAction.