AD-Cluster: Augmented Discriminative Clustering for Domain Adaptive Person Re-identification

Jie Chen

Abstract: Domain adaptive person re-identification (re-ID) is a challenging task, especially when person identities in target domains are unknown. Existing methods attempt to address this challenge by transferring image styles or aligning feature distributions across domains, whereas the rich unlabeled samples in target domains are not sufficiently exploited. This paper presents a novel augmented discriminative clustering (AD-Cluster) technique that estimates and augments person clusters in target domains and enforces the discrimination ability of re-ID models with the augmented clusters. AD-Cluster is trained by iterative density-based clustering, adaptive sample augmentation, and discriminative feature learning. It learns an image generator and a feature encoder that target to maximize intra-cluster distance (i.e., increase the diversity of sample space) and minimize intra-cluster distance in feature space (i.e., decrease the distance in new feature space), respectively. The key is to learn to increase the density of correctly predicted identities and the discrimination capability of re-ID models in an adversarial min-max manner. Extensive experiments over datasets Market-1501 and Duke MTMC-reID show that AD-Cluster outperforms the state-of-the-art with large margins.