Joint Visual and Temporal Consistency for Unsupervised Domain Adaptive Person Re-Identification

TL;DR

This paper introduces a novel unsupervised domain adaptive person re-identification method that jointly enforces visual and temporal consistency using local and global classification models, leading to improved performance.

Contribution

It proposes a unified framework combining Self-Adaptive Classification and Memory-based Temporal-guided Clustering for better label prediction and feature learning in unsupervised ReID.

Findings

Outperforms recent unsupervised domain adaptive methods on large-scale datasets.

Effectively leverages unlabeled data for discriminative feature learning.

Demonstrates superior results in both unsupervised and domain adaptive ReID tasks.

Abstract

Unsupervised domain adaptive person Re-IDentification (ReID) is challenging because of the large domain gap between source and target domains, as well as the lackage of labeled data on the target domain. This paper tackles this challenge through jointly enforcing visual and temporal consistency in the combination of a local one-hot classification and a global multi-class classification. The local one-hot classification assigns images in a training batch with different person IDs, then adopts a Self-Adaptive Classification (SAC) model to classify them. The global multi-class classification is achieved by predicting labels on the entire unlabeled training set with the Memory-based Temporal-guided Cluster (MTC). MTC predicts multi-class labels by considering both visual similarity and temporal consistency to ensure the quality of label prediction. The two classification models are combined in a unified framework, which effectively leverages the unlabeled data for discriminative feature learning. Experimental results on three large-scale ReID datasets demonstrate the superiority of proposed method in both unsupervised and unsupervised domain adaptive ReID tasks. For example, under unsupervised setting, our method outperforms recent unsupervised domain adaptive methods, which leverage more labels for training.