{"title":"An Improved Person Re-Identification Method based on AlignedReID ++ algorithm","authors":"Xiangyuan Zhu, Xiaozhou Dong, Hong Nie, Yusen Cen","doi":"10.1109/ICSAI57119.2022.10005320","DOIUrl":null,"url":null,"abstract":"Person re-identification (ReID) is a popular research topic in computer vision. It focuses on matching a given person from an image dataset captured by many non-overlapping cameras. It remains challenging duo to the influences of pose, illumination, occlusion, and background confusion. In this paper, an improved ReID approach based on the AlignedReID ++ algorithm is proposed. Three effective training tricks are introduced to improve the effectiveness of the AlignedReID ++ algorithm. Training loss, accuracy, and mean average precision (mAP) are used as measure metrics. Extensive experiments are implemented on the ResNet50 and DenseNet121 backbone networks. Our implementation gains the Rank-1 accuracy and mAP of 93.7% and 91.2%, respectively. The source code of the improved AlignReID ++ method is available on request.","PeriodicalId":339547,"journal":{"name":"2022 8th International Conference on Systems and Informatics (ICSAI)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 8th International Conference on Systems and Informatics (ICSAI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSAI57119.2022.10005320","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Person re-identification (ReID) is a popular research topic in computer vision. It focuses on matching a given person from an image dataset captured by many non-overlapping cameras. It remains challenging duo to the influences of pose, illumination, occlusion, and background confusion. In this paper, an improved ReID approach based on the AlignedReID ++ algorithm is proposed. Three effective training tricks are introduced to improve the effectiveness of the AlignedReID ++ algorithm. Training loss, accuracy, and mean average precision (mAP) are used as measure metrics. Extensive experiments are implemented on the ResNet50 and DenseNet121 backbone networks. Our implementation gains the Rank-1 accuracy and mAP of 93.7% and 91.2%, respectively. The source code of the improved AlignReID ++ method is available on request.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
