Classification Matters: Improving Video Action Detection with Class-Specific Attention
This work improves video action detection for applications like surveillance and video analysis, but it is incremental as it builds on existing attention-based methods.
The paper tackles the problem of video action detection by addressing classification bias towards actors, proposing a class-specific attention mechanism to focus on relevant contextual information. The result is superior performance on three benchmarks with significantly fewer parameters and less computation.
Video action detection (VAD) aims to detect actors and classify their actions in a video. We figure that VAD suffers more from classification rather than localization of actors. Hence, we analyze how prevailing methods form features for classification and find that they prioritize actor regions, yet often overlooking the essential contextual information necessary for accurate classification. Accordingly, we propose to reduce the bias toward actor and encourage paying attention to the context that is relevant to each action class. By assigning a class-dedicated query to each action class, our model can dynamically determine where to focus for effective classification. The proposed model demonstrates superior performance on three challenging benchmarks with significantly fewer parameters and less computation.