Adaptive Mutual Supervision for Weakly-Supervised Temporal Action Localization

TL;DR

This paper introduces an adaptive mutual supervision framework for weakly-supervised temporal action localization, effectively addressing the incompleteness of class activation sequences and improving localization accuracy through iterative mutual enhancement.

Contribution

The proposed AMS framework innovatively combines two branches with adaptive sampling and mutual pseudo-label supervision to better localize complete action regions.

Findings

Significantly outperforms state-of-the-art on THUMOS14.

Effective in localizing both discriminative and less discriminative action regions.

Iterative mutual supervision enhances localization completeness.

Abstract

Weakly-supervised temporal action localization aims to localize actions in untrimmed videos with only video-level action category labels. Most of previous methods ignore the incompleteness issue of Class Activation Sequences (CAS), suffering from trivial localization results. To solve this issue, we introduce an adaptive mutual supervision framework (AMS) with two branches, where the base branch adopts CAS to localize the most discriminative action regions, while the supplementary branch localizes the less discriminative action regions through a novel adaptive sampler. The adaptive sampler dynamically updates the input of the supplementary branch with a sampling weight sequence negatively correlated with the CAS from the base branch, thereby prompting the supplementary branch to localize the action regions underestimated by the base branch. To promote mutual enhancement between these two branches, we construct mutual location supervision. Each branch leverages location pseudo-labels generated from the other branch as localization supervision. By alternately optimizing the two branches in multiple iterations, we progressively complete action regions. Extensive experiments on THUMOS14 and ActivityNet1.2 demonstrate that the proposed AMS method significantly outperforms the state-of-the-art methods.