DASK: Distribution Rehearsing via Adaptive Style Kernel Learning for Exemplar-Free Lifelong Person Re-Identification

TL;DR

This paper introduces DASK, a novel exemplar-free lifelong person re-identification method that models and rehearses old domain distributions through adaptive style kernel learning, significantly reducing forgetting without storing old data.

Contribution

DASK proposes a new distribution rehearsing paradigm with adaptive style kernel learning, enhancing knowledge retention in lifelong person re-identification without exemplars.

Findings

Outperforms existing methods by 3.6%-6.8% in anti-forgetting

Achieves 4.5%-6.5% improvement in generalization capacity

Effective old-old knowledge consolidation without storing exemplars

Abstract

Lifelong person re-identification (LReID) is an important but challenging task that suffers from catastrophic forgetting due to significant domain gaps between training steps. Existing LReID approaches typically rely on data replay and knowledge distillation to mitigate this issue. However, data replay methods compromise data privacy by storing historical exemplars, while knowledge distillation methods suffer from limited performance due to the cumulative forgetting of undistilled knowledge. To overcome these challenges, we propose a novel paradigm that models and rehearses the distribution of the old domains to enhance knowledge consolidation during the new data learning, possessing a strong anti-forgetting capacity without storing any exemplars. Specifically, we introduce an exemplar-free LReID method called Distribution Rehearsing via Adaptive Style Kernel Learning (DASK). DASK includes a Distribution Rehearser Learning (DRL) mechanism that learns to transform arbitrary distribution data into the current data style at each learning step. To enhance the style transfer capacity of DRL, an Adaptive Kernel Prediction Network (AKPNet) is explored to achieve an instance-specific distribution adjustment. Additionally, we design a Distribution Rehearsing-driven LReID Training (DRRT) module, which rehearses old distribution based on the new data via the old AKPNet model, achieving effective new-old knowledge accumulation under a joint knowledge consolidation scheme. Experimental results show our DASK outperforms the existing methods by 3.6%-6.8% and 4.5%-6.5% on anti-forgetting and generalization capacity, respectively. Our code is available at https://github.com/zhoujiahuan1991/AAAI2025-LReID-DASK