Sub-action Prototype Learning for Point-level Weakly-supervised Temporal Action Localization

TL;DR

This paper introduces SPL-Loc, a novel framework for point-level weakly-supervised temporal action localization that leverages sub-action prototypes and alignment to improve pseudo label quality and boundary detection.

Contribution

The paper proposes a new sub-action prototype learning framework with clustering and alignment components, enhancing pseudo label generation for better localization accuracy.

Findings

Outperforms existing state-of-the-art methods on three benchmarks.

Effectively captures sub-action structures and temporal variations.

Improves action boundary prediction accuracy.

Abstract

Point-level weakly-supervised temporal action localization (PWTAL) aims to localize actions with only a single timestamp annotation for each action instance. Existing methods tend to mine dense pseudo labels to alleviate the label sparsity, but overlook the potential sub-action temporal structures, resulting in inferior performance. To tackle this problem, we propose a novel sub-action prototype learning framework (SPL-Loc) which comprises Sub-action Prototype Clustering (SPC) and Ordered Prototype Alignment (OPA). SPC adaptively extracts representative sub-action prototypes which are capable to perceive the temporal scale and spatial content variation of action instances. OPA selects relevant prototypes to provide completeness clue for pseudo label generation by applying a temporal alignment loss. As a result, pseudo labels are derived from alignment results to improve action boundary prediction. Extensive experiments on three popular benchmarks demonstrate that the proposed SPL-Loc significantly outperforms existing SOTA PWTAL methods.