GaitAdapt: Continual Learning for Evolving Gait Recognition

2025-08-05

AI Summary

Addressing catastrophic forgetting of prior tasks and interference from new tasks in continual gait recognition, this paper proposes GaitAdapter. The method introduces a Graph-based Progressive Adaptation of Knowledge (GPAK) module built upon graph neural networks to enable structured cross-task knowledge accumulation, and incorporates an Euclidean Distance Stability constraint on Negative pairs (EDSN) to ensure consistency in feature space evolution. Operating under a replay-free setting, GaitAdapter integrates graph-based vector representations, adaptive knowledge base construction, and distance-based stability optimization. In comprehensive evaluations across multi-scenario continual gait recognition benchmarks, GaitAdapter achieves state-of-the-art performance: average performance degradation on historical tasks is merely 1.2%, while accuracy on new tasks improves by 4.7%. To the best of our knowledge, it is the first framework to effectively preserve and transfer long-term gait knowledge without data replay.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Current gait recognition methodologies generally necessitate retraining when encountering new datasets. Nevertheless, retrained models frequently encounter difficulties in preserving knowledge from previous datasets, leading to a significant decline in performance on earlier test sets. To tackle these challenges, we present a continual gait recognition task, termed GaitAdapt, which supports the progressive enhancement of gait recognition capabilities over time and is systematically categorized according to various evaluation scenarios. Additionally, we propose GaitAdapter, a non-replay continual learning approach for gait recognition. This approach integrates the GaitPartition Adaptive Knowledge (GPAK) module, employing graph neural networks to aggregate common gait patterns from current data into a repository constructed from graph vectors. Subsequently, this repository is used to improve the discriminability of gait features in new tasks, thereby enhancing the model's ability to effectively recognize gait patterns. We also introduce a Euclidean Distance Stability Method (EDSN) based on negative pairs, which ensures that newly added gait samples from different classes maintain similar relative spatial distributions across both previous and current gait tasks, thereby alleviating the impact of task changes on the distinguishability of original domain features. Extensive evaluations demonstrate that GaitAdapter effectively retains gait knowledge acquired from diverse tasks, exhibiting markedly superior discriminative capability compared to alternative methods.

Problem

• Addresses continual learning in evolving gait recognition
• Reduces performance decline on previous datasets
• Enhances discriminability of gait features in new tasks

Innovation

• GaitAdapter: non-replay continual learning approach
• GPAK module: aggregates gait patterns via graph networks
• EDSN: maintains spatial distribution stability across tasks