Deep Transfer Learning for Person Re-identification

TL;DR

This paper introduces new deep transfer learning models for person re-identification that effectively address data scarcity, outperforming existing models on multiple benchmark datasets with both supervised and unsupervised approaches.

Contribution

It proposes a novel deep network architecture, a two-step fine-tuning strategy, and an unsupervised co-training model tailored for person Re-ID with limited data.

Findings

Achieved Rank-1 accuracy of 85.4% on CUHK03

Achieved Rank-1 accuracy of 83.7% on Market1501

Unsupervised model outperforms many supervised models on VIPeR

Abstract

Person re-identification (Re-ID) poses a unique challenge to deep learning: how to learn a deep model with millions of parameters on a small training set of few or no labels. In this paper, a number of deep transfer learning models are proposed to address the data sparsity problem. First, a deep network architecture is designed which differs from existing deep Re-ID models in that (a) it is more suitable for transferring representations learned from large image classification datasets, and (b) classification loss and verification loss are combined, each of which adopts a different dropout strategy. Second, a two-stepped fine-tuning strategy is developed to transfer knowledge from auxiliary datasets. Third, given an unlabelled Re-ID dataset, a novel unsupervised deep transfer learning model is developed based on co-training. The proposed models outperform the state-of-the-art deep Re-ID models by large margins: we achieve Rank-1 accuracy of 85.4\%, 83.7\% and 56.3\% on CUHK03, Market1501, and VIPeR respectively, whilst on VIPeR, our unsupervised model (45.1\%) beats most supervised models.