Towards Good Practices on Building Effective CNN Baseline Model for Person Re-identification

TL;DR

This paper proposes three practical guidelines for constructing effective CNN baseline models for person re-identification, significantly improving performance without complex domain-specific techniques.

Contribution

It introduces three specific practices—adding batch normalization, using a single fully-connected layer for identity classification, and adopting Adam optimizer—that enhance CNN baseline effectiveness.

Findings

Achieved state-of-the-art results on three benchmark datasets.

Demonstrated the effectiveness of simple architectural and training adjustments.

Facilitated better baseline models for person re-identification tasks.

Abstract

Person re-identification is indeed a challenging visual recognition task due to the critical issues of human pose variation, human body occlusion, camera view variation, etc. To address this, most of the state-of-the-art approaches are proposed based on deep convolutional neural network (CNN), being leveraged by its strong feature learning power and classification boundary fitting capacity. Although the vital role towards person re-identification, how to build effective CNN baseline model has not been well studied yet. To answer this open question, we propose 3 good practices in this paper from the perspectives of adjusting CNN architecture and training procedure. In particular, they are adding batch normalization after the global pooling layer, executing identity categorization directly using only one fully-connected, and using Adam as optimizer. The extensive experiments on 3 widely-used benchmark datasets demonstrate that, our propositions essentially facilitate the CNN baseline model to achieve the state-of-the-art performance without any other high-level domain knowledge or low-level technical trick.