Aligned Divergent Pathways for Omni-Domain Generalized Person Re-Identification

TL;DR

This paper introduces a novel multi-branch architecture called Aligned Divergent Pathways (ADP) with specialized modules to enhance omni-domain generalization in Person Re-Identification, outperforming existing methods across multiple benchmarks.

Contribution

It proposes ADP, a multi-branch backbone architecture with DyMAIN, PMoC, and DCML modules, to achieve effective omni-domain generalization in Person ReID, unifying multiple training and testing scenarios.

Findings

ADP surpasses state-of-the-art multi-source domain generalization methods.

ADP improves performance on single-source domain generalization benchmarks.

The proposed modules enhance feature robustness across diverse domains.

Abstract

Person Re-identification (Person ReID) has advanced significantly in fully supervised and domain generalized Person R e ID. However, methods developed for one task domain transfer poorly to the other. An ideal Person ReID method should be effective regardless of the number of domains involved in training or testing. Furthermore, given training data from the target domain, it should perform at least as well as state-of-the-art (SOTA) fully supervised Person ReID methods. We call this paradigm Omni-Domain Generalization Person ReID, referred to as ODG-ReID, and propose a way to achieve this by expanding compatible backbone architectures into multiple diverse pathways. Our method, Aligned Divergent Pathways (ADP), first converts a base architecture into a multi-branch structure by copying the tail of the original backbone. We design our module Dynamic Max-Deviance Adaptive Instance Normalization (DyMAIN) that encourages learning of generalized features that are robust to omni-domain directions and apply DyMAIN to the branches of ADP. Our proposed Phased Mixture-of-Cosines (PMoC) coordinates a mix of stable and turbulent learning rate schedules among branches for further diversified learning. Finally, we realign the feature space between branches with our proposed Dimensional Consistency Metric Loss (DCML). ADP outperforms the state-of-the-art (SOTA) results for multi-source domain generalization and supervised ReID within the same domain. Furthermore, our method demonstrates improvement on a wide range of single-source domain generalization benchmarks, achieving Omni-Domain Generalization over Person ReID tasks.