Few-shot Action Recognition with Prototype-centered Attentive Learning

TL;DR

This paper introduces a novel prototype-centered attentive learning approach for few-shot action recognition, addressing data efficiency and outlier issues, leading to significant performance improvements on standard benchmarks.

Contribution

The paper proposes a new PAL model with a prototype-centered contrastive loss and hybrid attentive mechanism, enhancing few-shot action recognition performance.

Findings

Outperforms previous state-of-the-art methods on four benchmarks.

Achieves over 10% improvement on fine-grained action recognition.

Effectively handles outliers and inter-class overlap.

Abstract

Few-shot action recognition aims to recognize action classes with few training samples. Most existing methods adopt a meta-learning approach with episodic training. In each episode, the few samples in a meta-training task are split into support and query sets. The former is used to build a classifier, which is then evaluated on the latter using a query-centered loss for model updating. There are however two major limitations: lack of data efficiency due to the query-centered only loss design and inability to deal with the support set outlying samples and inter-class distribution overlapping problems. In this paper, we overcome both limitations by proposing a new Prototype-centered Attentive Learning (PAL) model composed of two novel components. First, a prototype-centered contrastive learning loss is introduced to complement the conventional query-centered learning objective, in order to make full use of the limited training samples in each episode. Second, PAL further integrates a hybrid attentive learning mechanism that can minimize the negative impacts of outliers and promote class separation. Extensive experiments on four standard few-shot action benchmarks show that our method clearly outperforms previous state-of-the-art methods, with the improvement particularly significant (10+\%) on the most challenging fine-grained action recognition benchmark.