Distilling Vision-Language Pre-training to Collaborate with Weakly-Supervised Temporal Action Localization

TL;DR

This paper introduces a novel distillation framework that combines vision-language pre-training with classification-based pre-training to improve weakly-supervised temporal action localization, addressing the incomplete localization issue.

Contribution

It proposes a dual-branch distillation approach that fuses complementary knowledge from VLP and CBP to enhance action localization accuracy without extra annotations.

Findings

Significant performance improvements on THUMOS14 and ActivityNet1.2 datasets.

Effective fusion of VLP and CBP knowledge boosts localization accuracy.

A novel dual-branch training strategy enhances weakly-supervised learning.

Abstract

Weakly-supervised temporal action localization (WTAL) learns to detect and classify action instances with only category labels. Most methods widely adopt the off-the-shelf Classification-Based Pre-training (CBP) to generate video features for action localization. However, the different optimization objectives between classification and localization, make temporally localized results suffer from the serious incomplete issue. To tackle this issue without additional annotations, this paper considers to distill free action knowledge from Vision-Language Pre-training (VLP), since we surprisingly observe that the localization results of vanilla VLP have an over-complete issue, which is just complementary to the CBP results. To fuse such complementarity, we propose a novel distillation-collaboration framework with two branches acting as CBP and VLP respectively. The framework is optimized through a dual-branch alternate training strategy. Specifically, during the B step, we distill the confident background pseudo-labels from the CBP branch; while during the F step, the confident foreground pseudo-labels are distilled from the VLP branch. And as a result, the dual-branch complementarity is effectively fused to promote a strong alliance. Extensive experiments and ablation studies on THUMOS14 and ActivityNet1.2 reveal that our method significantly outperforms state-of-the-art methods.