Handling Label Uncertainty for Camera Incremental Person Re-Identification

TL;DR

This paper introduces ExtendOVA, a novel framework for camera incremental person re-identification that effectively handles class overlap, privacy constraints, and cross-camera relationships, significantly improving performance over existing methods.

Contribution

The paper proposes ExtendOVA, a new approach that addresses class overlap and privacy issues in incremental person ReID by using instance-wise identification, pseudo-label correction, and prototypical memory.

Findings

Outperforms state-of-the-art methods on multiple benchmarks.

Effectively handles class overlap and privacy constraints.

Improves inter-camera relationship retention.

Abstract

Incremental learning for person re-identification (ReID) aims to develop models that can be trained with a continuous data stream, which is a more practical setting for real-world applications. However, the existing incremental ReID methods make two strong assumptions that the cameras are fixed and the new-emerging data is class-disjoint from previous classes. This is unrealistic as previously observed pedestrians may re-appear and be captured again by new cameras. In this paper, we investigate person ReID in an unexplored scenario named Camera Incremental Person ReID (CIPR), which advances existing lifelong person ReID by taking into account the class overlap issue. Specifically, new data collected from new cameras may probably contain an unknown proportion of identities seen before. This subsequently leads to the lack of cross-camera annotations for new data due to privacy concerns. To address these challenges, we propose a novel framework ExtendOVA. First, to handle the class overlap issue, we introduce an instance-wise seen-class identification module to discover previously seen identities at the instance level. Then, we propose a criterion for selecting confident ID-wise candidates and also devise an early learning regularization term to correct noise issues in pseudo labels. Furthermore, to compensate for the lack of previous data, we resort prototypical memory bank to create surrogate features, along with a cross-camera distillation loss to further retain the inter-camera relationship. The comprehensive experimental results on multiple benchmarks show that ExtendOVA significantly outperforms the state-of-the-arts with remarkable advantages.