Domain-Class Correlation Decomposition for Generalizable Person Re-Identification
Abstract
Domain generalization in person re-identification is a highly important meaningful and practical task in which a model trained with data from several source domains is expected to generalize well to unseen target domains. Domain adversarial learning is a promising domain generalization method that aims to remove domain information in the latent representation through adversarial training. However, in person re-identification, the domain and class are correlated, and we theoretically show that domain adversarial learning will lose certain information about class due to this domain-class correlation. Inspired by casual inference, we propose to perform interventions to the domain factor , aiming to decompose the domain-class correlation. To achieve this goal, we proposed estimating the resulting representation caused by the intervention through first- and second-order statistical characteristic matching. Specifically, we build a memory bank to restore the statistical characteristics of each domain. Then, we use the newly generated samples to compute the loss function. These samples are domain-class correlation decomposed; thus, we can learn a domain-invariant representation that can capture more class-related features. Extensive experiments show that our model outperforms the state-of-the-art methods on the large-scale domain generalization Re-ID benchmark.
Cite
@article{arxiv.2106.15206,
title = {Domain-Class Correlation Decomposition for Generalizable Person Re-Identification},
author = {Kaiwen Yang and Xinmei Tian},
journal= {arXiv preprint arXiv:2106.15206},
year = {2021}
}
Comments
10 pages, 5 figures