Pose-Transformation and Radial Distance Clustering for Unsupervised Person Re-identification

TL;DR

This paper presents an unsupervised person re-identification method that uses pose transformations and radial distance clustering to improve feature discrimination without requiring labeled data.

Contribution

It introduces a novel two-stage training strategy with pose augmentation and a radial distance loss for effective unsupervised re-ID.

Findings

Outperforms state-of-the-art unsupervised re-ID methods on large datasets.

Effective in creating compact, well-separated feature clusters.

Demonstrates robustness across multiple datasets.

Abstract

Person re-identification (re-ID) aims to tackle the problem of matching identities across non-overlapping cameras. Supervised approaches require identity information that may be difficult to obtain and are inherently biased towards the dataset they are trained on, making them unscalable across domains. To overcome these challenges, we propose an unsupervised approach to the person re-ID setup. Having zero knowledge of true labels, our proposed method enhances the discriminating ability of the learned features via a novel two-stage training strategy. The first stage involves training a deep network on an expertly designed pose-transformed dataset obtained by generating multiple perturbations for each original image in the pose space. Next, the network learns to map similar features closer in the feature space using the proposed discriminative clustering algorithm. We introduce a novel radial distance loss, that attends to the fundamental aspects of feature learning - compact clusters with low intra-cluster and high inter-cluster variation. Extensive experiments on several large-scale re-ID datasets demonstrate the superiority of our method compared to state-of-the-art approaches.