Margin CosReid Network for Pedestrian Re-Identification

Xiao Yun,Min Ge,Xiaofeng Hou,Yanjing Sun,Kaiwen Dong

doi:10.3390/app11041775

Abstract

This paper proposes a margin CosReid network for effective pedestrian re-identification. Aiming to overcome the overfitting, gradient explosion, and loss function non-convergence problems caused by traditional CNNs, the proposed GBNeck model can realize a faster, stronger generalization, and more discriminative feature extraction task. Furthermore, to enhance the classification ability of the softmax loss function within classes, the margin cosine softmax loss (MCSL) is proposed through a boundary margin introduction to ensure intraclass compactness and interclass separability of the learning depth features and thus to build a stronger metric-based learning model for pedestrian re-identification. The effectiveness of the margin CosReid network was verified on the mainstream datasets Market-1501 and DukeMTMC-reID compared with other state-of-the-art pedestrian re-identification methods.

Highlights

Given a probe image, the goal of pedestrian re-identification is to search for images containing the same person in a gallery under multiple nonoverlapping cameras [1]
This paper proposes the margin cosine softmax loss (MCSL) to normalize the weight and feature vector and introduces a boundary margin parameter m for maximizing the difference between classes while minimizing those within classes to embed the pedestrian features deeper
We demonstrate the effectiveness of the proposed method by using the cumulative matching characteristics (CMC) at rank-1, rank-5, and rank-10 and the mean average precision on the standard dataset

Summary

Introduction

Given a probe image (query), the goal of pedestrian re-identification is to search for images containing the same person in a gallery (database with training labels) under multiple nonoverlapping cameras [1]. It has played a supervisory role in public video-based pedestrian monitoring, the performance can be seriously affected by the presence or absence of obstructions and changes in target posture, camera angles, and illumination intensity. The convolutional neural network (CNN), one of the typical feature extraction methods in deep learning, embeds the target feature space from the dataset automatically in the pedestrian re-identification problem. Problems of overfitting, gradient explosion, and loss function non-convergence may occur in these networks, which are not very effective in dealing with complex re-identification tasks

Methods

Results

Conclusion