Distance learning by mining hard and easy negative samples for person re-identification. (November 2019)

• Record Type: Journal Article
• Authors: Zhu, Xiaoke; Jing, Xiao-Yuan; Zhang, Fan; Zhang, Xinyu; You, Xinge; Cui, Xiang
• Is Part Of: Pattern recognition. Volume 95(2019:Nov.)
• Page Start: 211
• Page End: 222
• Publication Date: 2019-11
• Subjects: Distance learning -- Symmetric triplet constraint -- Negative samples division -- Projection matrix -- Person re-identification; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.06.007
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We have proposed a Hard and Easy Negative samples mining based Distance learning (HEND) approach for person re-identification. We have designed a symmetric triplet constraint for the proposed HEND approach. We have proposed a Projection based HEND (PHEND) approach, which simultaneously learns a projection matrix and a distance metric. We have conducted extensive experiments in this paper to evaluate our approaches. Abstract: Distance learning is an effective technique for person re-identification. In practice, the hard negative samples usually contain more discriminative information than the easy negative samples. Therefore, it's necessary to investigate how to make full use of the discriminative information conveyed by different types of negative samples in the distance learning process. In this paper, we propose a H ard and E asy N egative samples mining based D istance learning (HEND) approach for person re-identification, which learns the distance metric by designing different objective functions for hard and easy negative samples, such that the discriminative information contained in negative samples can be exploited more effectively. Moreover, considering that there usually exist large differences between the images captured by different cameras, we further propose a projection-based HEND approach to reduce the influence of between-camera differences to the re-identification. Experimental results on seven pedestrian image datasets demonstrate the effectiveness of the proposed approaches.