Cross-video Identity Correlating for Person Re-identification Pre-training

TL;DR

This paper introduces CION, a novel pre-training framework for person re-identification that leverages cross-video identity correlation and self-distillation to improve performance with fewer samples.

Contribution

The paper proposes a new cross-video identity correlating pre-training method with a multi-level denoising approach and identity-guided self-distillation, advancing person re-identification models.

Findings

CION achieves higher mAP and rank-1 accuracy on Market1501 and MSMT17 datasets.

CION outperforms previous state-of-the-art with fewer training samples.

A diverse model zoo with 32 pre-trained models is provided.

Abstract

Recent researches have proven that pre-training on large-scale person images extracted from internet videos is an effective way in learning better representations for person re-identification. However, these researches are mostly confined to pre-training at the instance-level or single-video tracklet-level. They ignore the identity-invariance in images of the same person across different videos, which is a key focus in person re-identification. To address this issue, we propose a Cross-video Identity-cOrrelating pre-traiNing (CION) framework. Defining a noise concept that comprehensively considers both intra-identity consistency and inter-identity discrimination, CION seeks the identity correlation from cross-video images by modeling it as a progressive multi-level denoising problem. Furthermore, an identity-guided self-distillation loss is proposed to implement better large-scale pre-training by mining the identity-invariance within person images. We conduct extensive experiments to verify the superiority of our CION in terms of efficiency and performance. CION achieves significantly leading performance with even fewer training samples. For example, compared with the previous state-of-the-art~\cite{ISR}, CION with the same ResNet50-IBN achieves higher mAP of 93.3\% and 74.3\% on Market1501 and MSMT17, while only utilizing 8\% training samples. Finally, with CION demonstrating superior model-agnostic ability, we contribute a model zoo named ReIDZoo to meet diverse research and application needs in this field. It contains a series of CION pre-trained models with spanning structures and parameters, totaling 32 models with 10 different structures, including GhostNet, ConvNext, RepViT, FastViT and so on. The code and models will be made publicly available at https://github.com/Zplusdragon/CION_ReIDZoo.