Forcing the Whole Video as Background: An Adversarial Learning Strategy for Weakly Temporal Action Localization
This work addresses a specific challenge in video analysis for researchers in computer vision, offering an incremental improvement over existing methods by reducing reliance on artificial assumptions for background modeling.
The paper tackles the problem of weakly supervised temporal action localization by addressing mis-activation of background snippets, proposing an adversarial learning strategy that forces the whole video as background to enhance action modeling, achieving improved performance on THUMOS14 and ActivityNet1.2 benchmarks.
With video-level labels, weakly supervised temporal action localization (WTAL) applies a localization-by-classification paradigm to detect and classify the action in untrimmed videos. Due to the characteristic of classification, class-specific background snippets are inevitably mis-activated to improve the discriminability of the classifier in WTAL. To alleviate the disturbance of background, existing methods try to enlarge the discrepancy between action and background through modeling background snippets with pseudo-snippet-level annotations, which largely rely on artificial hypotheticals. Distinct from the previous works, we present an adversarial learning strategy to break the limitation of mining pseudo background snippets. Concretely, the background classification loss forces the whole video to be regarded as the background by a background gradient reinforcement strategy, confusing the recognition model. Reversely, the foreground(action) loss guides the model to focus on action snippets under such conditions. As a result, competition between the two classification losses drives the model to boost its ability for action modeling. Simultaneously, a novel temporal enhancement network is designed to facilitate the model to construct temporal relation of affinity snippets based on the proposed strategy, for further improving the performance of action localization. Finally, extensive experiments conducted on THUMOS14 and ActivityNet1.2 demonstrate the effectiveness of the proposed method.