Intra-Camera Supervised Person Re-Identification

TL;DR

This paper introduces a scalable person re-identification approach that reduces annotation effort by using intra-camera labels and a multi-task deep learning framework to discover cross-camera identities, achieving high accuracy.

Contribution

It proposes the Intra-Camera Supervised (ICS) re-id paradigm and the MATE deep learning method for cross-camera identity matching without inter-camera labels.

Findings

MATE achieves 88.7% rank-1 on Market-1501 in ICS setting.

Outperforms unsupervised methods significantly.

Closely approaches fully supervised performance.

Abstract

Existing person re-identification (re-id) methods mostly exploit a large set of cross-camera identity labelled training data. This requires a tedious data collection and annotation process, leading to poor scalability in practical re-id applications. On the other hand unsupervised re-id methods do not need identity label information, but they usually suffer from much inferior and insufficient model performance. To overcome these fundamental limitations, we propose a novel person re-identification paradigm based on an idea of independent per-camera identity annotation. This eliminates the most time-consuming and tedious inter-camera identity labelling process, significantly reducing the amount of human annotation efforts. Consequently, it gives rise to a more scalable and more feasible setting, which we call Intra-Camera Supervised (ICS) person re-id, for which we formulate a Multi-tAsk mulTi-labEl (MATE) deep learning method. Specifically, MATE is designed for self-discovering the cross-camera identity correspondence in a per-camera multi-task inference framework. Extensive experiments demonstrate the cost-effectiveness superiority of our method over the alternative approaches on three large person re-id datasets. For example, MATE yields 88.7% rank-1 score on Market-1501 in the proposed ICS person re-id setting, significantly outperforming unsupervised learning models and closely approaching conventional fully supervised learning competitors.