Boosting Few-shot Action Recognition with Graph-guided Hybrid Matching

TL;DR

This paper introduces GgHM, a novel framework for few-shot action recognition that leverages graph-guided prototype construction and hybrid matching strategies to improve recognition accuracy, especially among similar categories.

Contribution

The paper proposes a new graph-guided hybrid matching framework that explicitly models class prototype relationships and combines frame- and tuple-level matching for better few-shot action recognition.

Findings

GgHM outperforms existing baselines on multiple datasets.

The graph-guided prototype construction improves class discrimination.

Hybrid matching enhances recognition of similar categories.

Abstract

Class prototype construction and matching are core aspects of few-shot action recognition. Previous methods mainly focus on designing spatiotemporal relation modeling modules or complex temporal alignment algorithms. Despite the promising results, they ignored the value of class prototype construction and matching, leading to unsatisfactory performance in recognizing similar categories in every task. In this paper, we propose GgHM, a new framework with Graph-guided Hybrid Matching. Concretely, we learn task-oriented features by the guidance of a graph neural network during class prototype construction, optimizing the intra- and inter-class feature correlation explicitly. Next, we design a hybrid matching strategy, combining frame-level and tuple-level matching to classify videos with multivariate styles. We additionally propose a learnable dense temporal modeling module to enhance the video feature temporal representation to build a more solid foundation for the matching process. GgHM shows consistent improvements over other challenging baselines on several few-shot datasets, demonstrating the effectiveness of our method. The code will be publicly available at https://github.com/jiazheng-xing/GgHM.