CO-EVO: Co-evolving Semantic Anchoring and Style Diversification for Federated DG-ReID

2026-04-29

AI Summary

This work addresses the challenge of model overfitting and poor generalization to unseen domains in federated domain generalization person re-identification, caused by heterogeneous camera styles across clients. To tackle this issue, the authors propose a semantic-and-style co-evolution framework that decouples identity-relevant features from domain-specific biases while preserving data privacy. Specifically, Camera-invariant Semantic Anchoring (CSA) is introduced to purify identity semantics, while Global Style Diversification (GSD), driven by a Global Camera Style Bank (GCSB), enriches visual style variations. The synergistic interaction between semantic purification and style diversification significantly enhances cross-domain generalization. Extensive experiments demonstrate that the proposed method achieves state-of-the-art performance on multiple benchmarks for federated domain generalization person re-identification.

Abstract

Federated domain generalization for person re-identification (FedDG-ReID) aims to collaboratively train a pedestrian retrieval model across multiple decentralized source domains such that it can generalize to unseen target environments without compromising raw data privacy. However, this task is significantly challenged by the inherent stylistic gaps across decentralized clients. Without global supervision, models easily succumb to shortcut learning where representations overfit to domain specific camera biases rather than universal identity features. We propose CO-EVO, a novel federated framework that resolves this semantic-style conflict through a co-evolutionary mechanism. On the semantic side, Camera-Invariant Semantic Anchoring (CSA) learns identity prompts with cross-camera consistency to establish purified and domain-agnostic anchors that filter out local imaging noise. On the visual side, Global Style Diversification (GSD), powered by a Global Camera-Style Bank (GCSB), synthesizes realistic perturbations to expand the visual boundaries of training data. The core of CO-EVO is its co-evolutionary loop where purified anchors act as gravitational centers to guide the image encoder toward robust anatomical attributes amidst diverse style variations. Extensive experiments demonstrate that CO-EVO achieves state-of-the-art (SOTA) performance, proving that the synergy between semantic purification and style expansion is essential for robust cross-domain generalization. Our code is available at: https://github.com/NanYiyuzurn/ACL-LGPS-2026.

Problem

• Federated Domain Generalization
• Person Re-identification
• Style Gap
• Shortcut Learning
• Data Privacy

Innovation

• Federated Domain Generalization
• Semantic Anchoring
• Style Diversification
• Co-evolutionary Learning
• Person Re-identification