Leader-Based Multi-Scale Attention Deep Architecture for Person Re-Identification

Person re-identification (re-id) aims to match people across non-overlapping camera views in a public space. This is a challenging problem because the people captured in surveillance videos often wear similar clothing. Consequently, the differences in their appearance are typically subtle and only detectable at particular locations and scales. In this paper, we propose a deep re-id network (MuDeep) that is composed of two novel types of layers - a multi-scale deep learning layer, and a leader-based attention learning layer. Specifically, the former learns deep discriminative feature representations at different scales, while the latter utilizes the information from multiple scales to lead and determine the optimal weightings for each scale. The importance of different spatial locations for extracting discriminative features is learned explicitly via our leader-based attention learning layer. Extensive experiments are carried out to demonstrate that the proposed MuDeep outperforms the state-of-the-art on a number of benchmarks and has a better generalization ability under a domain generalization setting.

1 Introduction

Person re-identification (re-id) is the task of matching two pedestrian images crossing non-overlapping camera views [1]. It plays an important role in a number of applications in video surveillance, including multi-camera tracking [2], [3], crowd counting [4], [5], and multi-camera activity analysis [6], [7]. Person re-id is extremely challenging and remains unsolved for a number of reasons. First, in different camera views, one person's appearance often changes dramatically due to the variances in body poses, camera viewpoints, occlusion and illumination conditions. Second, in a public space, many people often wear very similar clothing (e.g., dark coats in winter). Thus, the differences that can be used to distinguish between people are subtle. These subtle discrepancies exist at different locations in the image and are of different spatial scales. For instance, they could be global, e.g., one person is bulkier than another, or local, e.g., the two people are wearing different shoes.