TAL: Two-stream Adaptive Learning for Generalizable Person Re-identification

TL;DR

This paper introduces TAL, a two-stream adaptive learning framework that effectively captures both domain-specific and domain-invariant features to improve generalization in person re-identification across unseen domains.

Contribution

The paper proposes a novel two-stream adaptive learning framework that models both domain-specific and domain-invariant features for better generalization in person re-id.

Findings

Outperforms state-of-the-art methods on domain generalization tasks.

Effective modeling of both domain-specific and invariant features.

Applicable to both single-source and multi-source scenarios.

Abstract

Domain generalizable person re-identification aims to apply a trained model to unseen domains. Prior works either combine the data in all the training domains to capture domain-invariant features, or adopt a mixture of experts to investigate domain-specific information. In this work, we argue that both domain-specific and domain-invariant features are crucial for improving the generalization ability of re-id models. To this end, we design a novel framework, which we name two-stream adaptive learning (TAL), to simultaneously model these two kinds of information. Specifically, a domain-specific stream is proposed to capture training domain statistics with batch normalization (BN) parameters, while an adaptive matching layer is designed to dynamically aggregate domain-level information. In the meantime, we design an adaptive BN layer in the domain-invariant stream, to approximate the statistics of various unseen domains. These two streams work adaptively and collaboratively to learn generalizable re-id features. Our framework can be applied to both single-source and multi-source domain generalization tasks, where experimental results show that our framework notably outperforms the state-of-the-art methods.