Multi-Domain Adversarial Feature Generalization for Person Re-Identification

TL;DR

This paper introduces a multi-dataset adversarial auto-encoder approach for person re-identification that learns a universal, domain-invariant feature representation capable of generalizing to unseen camera systems without requiring target domain data during training.

Contribution

The paper proposes MMFA-AAE, a novel multi-dataset adversarial auto-encoder that enhances person Re-ID by learning domain-invariant features for better generalization to new environments.

Findings

Outperforms most domain generalization methods in person Re-ID.

Surpasses many supervised and unsupervised domain adaptation methods.

Effectively learns universal features that generalize to unseen camera systems.

Abstract

With the assistance of sophisticated training methods applied to single labeled datasets, the performance of fully-supervised person re-identification (Person Re-ID) has been improved significantly in recent years. However, these models trained on a single dataset usually suffer from considerable performance degradation when applied to videos of a different camera network. To make Person Re-ID systems more practical and scalable, several cross-dataset domain adaptation methods have been proposed, which achieve high performance without the labeled data from the target domain. However, these approaches still require the unlabeled data of the target domain during the training process, making them impractical. A practical Person Re-ID system pre-trained on other datasets should start running immediately after deployment on a new site without having to wait until sufficient images or videos are collected and the pre-trained model is tuned. To serve this purpose, in this paper, we reformulate person re-identification as a multi-dataset domain generalization problem. We propose a multi-dataset feature generalization network (MMFA-AAE), which is capable of learning a universal domain-invariant feature representation from multiple labeled datasets and generalizing it to `unseen' camera systems. The network is based on an adversarial auto-encoder to learn a generalized domain-invariant latent feature representation with the Maximum Mean Discrepancy (MMD) measure to align the distributions across multiple domains. Extensive experiments demonstrate the effectiveness of the proposed method. Our MMFA-AAE approach not only outperforms most of the domain generalization Person Re-ID methods, but also surpasses many state-of-the-art supervised methods and unsupervised domain adaptation methods by a large margin.