Label-Efficient Skeleton-based Recognition with Stable-Invertible Graph Convolutional Networks

TL;DR

This paper introduces a label-efficient graph convolutional network approach for skeleton-based action recognition, leveraging invertible GCNs to select the most informative data subsets, reducing labeling costs while maintaining high accuracy.

Contribution

It proposes a novel acquisition function for selecting informative data subsets and extends it with invertible GCNs to better capture data distribution in latent space.

Findings

Outperforms existing methods on benchmark datasets

Reduces labeling requirements significantly

Demonstrates robustness across different datasets

Abstract

Skeleton-based action recognition is a hotspot in image processing. A key challenge of this task lies in its dependence on large, manually labeled datasets whose acquisition is costly and time-consuming. This paper devises a novel, label-efficient method for skeleton-based action recognition using graph convolutional networks (GCNs). The contribution of the proposed method resides in learning a novel acquisition function -- scoring the most informative subsets for labeling -- as the optimum of an objective function mixing data representativity, diversity and uncertainty. We also extend this approach by learning the most informative subsets using an invertible GCN which allows mapping data from ambient to latent spaces where the inherent distribution of the data is more easily captured. Extensive experiments, conducted on two challenging skeleton-based recognition datasets, show the effectiveness and the outperformance of our label-frugal GCNs against the related work.