Affinity Contrastive Learning for Skeleton-Based Human Activity Understanding

doi:10.1109/TBIOM.2026.3652637

Affinity Contrastive Learning for Skeleton-Based Human Activity Understanding

2026-03-01

0

PRE

AI

DOI：10.1109/TBIOM.2026.3652637

Original

Original request for help

Save

Abstract

En 中文

In skeleton-based human activity understanding, existing methods often adopt the contrastive learning paradigm to construct a discriminative feature space. However, many of these approaches fail to exploit the structural inter-class similarities and overlook the impact of anomalous positive samples. In this study, we introduce ACLNet, an Affinity Contrastive Learning Network that explores the intricate clustering relationships among human activity classes to improve feature discrimination. Specifically, we propose an affinity metric to refine similarity measurements, thereby forming activity superclasses that provide more informative contrastive signals. A dynamic temperature schedule is also introduced to adaptively adjust the penalty strength for various superclasses. In addition, we employ a margin-based contrastive strategy to improve the separation of hard positive and negative samples within classes. Extensive experiments on NTU RGB+D 60, NTU RGB+D 120, Kinetics-Skeleton, PKU-MMD, FineGYM, and CASIA-B demonstrate the superiority of our method in skeleton-based action recognition, gait recognition, and person re-identification. The source code is available at https://github.com/firework8/ACLNet

Keywords:

Contrastive learning

Skeleton

Biometrics

Semantics

Artificial intelligence

Temperature measurement

Measurement

Gait recognition

Feature extraction

Writing

Human activity understanding

skeleton

graph convolutional networks

contrastive learning

Affinity Contrastive Learning for Skeleton-Based Human Activity Understanding

Abstract

Journal

Organization