Rethinking Person Re-Identification via Semantic-Based Pretraining

TL;DR

This paper proposes a semantic-based pretraining method called VTBR for person re-identification, utilizing dense captions from a new dataset, which achieves competitive results with less data than traditional ImageNet pretraining.

Contribution

Introduces a novel semantic pretraining approach using dense captions and a new dataset, reducing data requirements for effective person Re-ID model initialization.

Findings

VTBR achieves comparable performance to ImageNet pretraining.

Uses up to 1.4x fewer images for pretraining.

Demonstrates the effectiveness of semantic pretraining in Re-ID tasks.

Abstract

Pretraining is a dominant paradigm in computer vision. Generally, supervised ImageNet pretraining is commonly used to initialize the backbones of person re-identification (Re-ID) models. However, recent works show a surprising result that CNN-based pretraining on ImageNet has limited impacts on Re-ID system due to the large domain gap between ImageNet and person Re-ID data. To seek an alternative to traditional pretraining, here we investigate semantic-based pretraining as another method to utilize additional textual data against ImageNet pretraining. Specifically, we manually construct a diversified FineGPR-C caption dataset for the first time on person Re-ID events. Based on it, a pure semantic-based pretraining approach named VTBR is proposed to adopt dense captions to learn visual representations with fewer images. We train convolutional neural networks from scratch on the captions of FineGPR-C dataset, and then transfer them to downstream Re-ID tasks. Comprehensive experiments conducted on benchmark datasets show that our VTBR can achieve competitive performance compared with ImageNet pretraining - despite using up to 1.4x fewer images, revealing its potential in Re-ID pretraining.