Discriminative fine-grained network for vehicle re-identification using two-stage re-ranking

作     者：Qi WANG Weidong MIN Daojing HE Song ZOU Tiemei HUANG Yu ZHANG Ruikang LIU Qi WANG;Weidong MIN;Daojing HE;Song ZOU;Tiemei HUANG;Yu ZHANG;Ruikang LIU

作者机构：School of Information Engineering Nanchang University School of Software Nanchang University Jiangxi Key Laboratory of Smart City Nanchang University School of Computer Science and Software Engineering East China Normal University 

出 版 物：《Science China(Information Sciences)》 (中国科学:信息科学(英文版))

年 卷 期：2020年第63卷第11期

页      面：147-158页

核心收录：

学科分类：0810[工学-信息与通信工程] 0808[工学-电气工程] 08[工学] 080203[工学-机械设计及理论] 0838[工学-公安技术] 0802[工学-机械工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by National Natural Science Foundation of China (Grant Nos. 61762061, 62076117) National Key R&D Program of China (Grant Nos. 2017YFB0801701, 2017YFB0802805) Natural Science Foundation of Jiangxi Province (Grant No. 20161ACB20004) Jiangxi Key Laboratory of Smart City (Grant No. 20192BCD40002) 

主　　题：vehicle re-identification DFN two-stage re-ranking fine-grained Jaccard metric 

摘      要：Research on the application of vehicle re-identification to video surveillance has attracted increasingly growing attention. Existing methods are associated with the difficulties of distinguishing different instances of the same car model owing to the incapability of recognizing subtle differences among these instances and the possibility that a subtle difference may lead to incorrect results of ranking. In this paper, a discriminative fine-grained network for vehicle re-identification based on a two-stage re-ranking framework is proposed to address these issues. This discriminative fine-grained network(DFN) is composed of fine-grained and Siamese networks. The proposed hybrid network can extract discriminative features of the vehicle instances with subtle differences. The Siamese network is rather suitable for general object re-identification using two streams of the network, while the fine-grained network is capable of detecting subtle *** proposed two-stage re-ranking method allows obtaining a more reliable ranking list by using the Jaccard metric and merging the first and second re-ranking lists, where the latter list is formed using the sample mean feature. Experimental results on the VeR i-776 and VehicleID datasets show that the proposed method achieves the superior performance compared to the state-of-the-art methods used in vehicle re-identification.