Weakly-Supervised Action Localization by Generative Attention Modeling

TL;DR

This paper introduces a novel weakly-supervised action localization method using a conditional Variational Auto-Encoder to better distinguish action frames from context, addressing the common action-context confusion issue.

Contribution

It proposes a probabilistic modeling approach with conditional VAE to improve action localization accuracy under weak supervision, a novel application in this domain.

Findings

Outperforms existing methods on THUMOS14 and ActivityNet1.2 datasets.

Effectively reduces action-context confusion in localization.

Demonstrates the benefit of probabilistic modeling for weakly-supervised learning.

Abstract

Weakly-supervised temporal action localization is a problem of learning an action localization model with only video-level action labeling available. The general framework largely relies on the classification activation, which employs an attention model to identify the action-related frames and then categorizes them into different classes. Such method results in the action-context confusion issue: context frames near action clips tend to be recognized as action frames themselves, since they are closely related to the specific classes. To solve the problem, in this paper we propose to model the class-agnostic frame-wise probability conditioned on the frame attention using conditional Variational Auto-Encoder (VAE). With the observation that the context exhibits notable difference from the action at representation level, a probabilistic model, i.e., conditional VAE, is learned to model the likelihood of each frame given the attention. By maximizing the conditional probability with respect to the attention, the action and non-action frames are well separated. Experiments on THUMOS14 and ActivityNet1.2 demonstrate advantage of our method and effectiveness in handling action-context confusion problem. Code is now available on GitHub.