Towards Precise Intra-camera Supervised Person Re-identification

TL;DR

This paper introduces a novel intra-camera supervision framework for person re-identification that reduces annotation effort and achieves performance comparable to fully supervised methods by combining intra-camera learning with inter-camera association.

Contribution

It proposes camera-specific classifiers and a hybrid loss for intra-camera learning, along with a graph-based inter-camera association, advancing ICS Re-ID performance.

Findings

Outperforms existing ICS methods significantly.

Achieves comparable results to fully supervised methods on two datasets.

Effective integration of intra-camera and inter-camera learning modules.

Abstract

Intra-camera supervision (ICS) for person re-identification (Re-ID) assumes that identity labels are independently annotated within each camera view and no inter-camera identity association is labeled. It is a new setting proposed recently to reduce the burden of annotation while expect to maintain desirable Re-ID performance. However, the lack of inter-camera labels makes the ICS Re-ID problem much more challenging than the fully supervised counterpart. By investigating the characteristics of ICS, this paper proposes camera-specific non-parametric classifiers, together with a hybrid mining quintuplet loss, to perform intra-camera learning. Then, an inter-camera learning module consisting of a graph-based ID association step and a Re-ID model updating step is conducted. Extensive experiments on three large-scale Re-ID datasets show that our approach outperforms all existing ICS works by a great margin. Our approach performs even comparable to state-of-the-art fully supervised methods in two of the datasets.