{"type":"ScholarlyArticle","title":"Self-Guided Hard Negative Generation for Unsupervised Person Re-Identification","authors":[{"name":"Dongdong Li","slug":"dongdong-li","given":"Dongdong","family":"Li","affiliation":"Institute of Automation , Chinese Academy of Sciences (CASIA) , China; University of Chinese Academy of Sciences (UCAS) , China"},{"name":"Zhigang Wang","slug":"zhigang-wang","given":"Zhigang","family":"Wang","affiliation":"Baidu VIS , China"},{"name":"Jian Wang","slug":"jian-wang","given":"Jian","family":"Wang","affiliation":"Baidu VIS , China"},{"name":"Xinyu Zhang","slug":"xinyu-zhang","given":"Xinyu","family":"Zhang","affiliation":"Baidu VIS , China"},{"name":"Errui Ding","slug":"errui-ding","given":"Errui","family":"Ding","affiliation":"Baidu VIS , China"},{"name":"Jingdong Wang","slug":"jingdong-wang","given":"Jingdong","family":"Wang","affiliation":"Baidu VIS , China"},{"name":"Zhaoxiang Zhang","slug":"zhaoxiang-zhang","given":"Zhaoxiang","family":"Zhang","affiliation":"Institute of Automation , Chinese Academy of Sciences (CASIA) , China; University of Chinese Academy of Sciences (UCAS) , China; Centre for Artificial Intelligence and Robotics , HKISI CAS , China"}],"abstract":"Recent unsupervised person re-identification (reID) methods mostly apply pseudo labels from clustering algorithms as supervision signals. Despite great success, this fashion is very likely to aggregate different identities with similar appearances into the same cluster. In result, the hard negative samples, playing important role in training reID models, are significantly reduced. To alleviate this problem, we propose a self-guided hard negative generation method for unsupervised person re-ID. Specifically, a joint framework is developed which incorporates a hard negative generation network (HNGN) and a re-ID network. To continuously generate harder negative samples to provide effective supervisions in the contrastive learning, the two networks are alternately trained in an adversarial manner to improve each other, where the reID network guides HNGN to generate challenging data and HNGN enforces the re-ID network to enhance discrimination ability. During inference, the performance of re-ID network is improved without introducing any extra parameters. Extensive experiments demonstrate that the proposed method significantly outperforms a strong baseline and also achieves better results than state-of-the-art methods.","doi":"10.24963/ijcai.2022/149","url":"https://ijcai.org/proceedings/2022/149","pdf":"https://ijcai.org/proceedings/2022/0149.pdf","markdown":"https://ijcai.org/proceedings/2022/149.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/149","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Computer Vision","track":"Main Track","number":149,"pages":{"first":1067,"last":1073},"keywords":["Computer Vision: Image and Video retrieval"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986798","length_sec":91,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984941","length_sec":354,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Bai et al., 2021] Zechen Bai, Zhigang Wang, Jian Wang, Di Hu, and Errui Ding. Unsupervised multi-source domain adaptation for person re-identification. In CVPR, 2021.","title":"Unsupervised multi-source domain adaptation for person re-identification","authors":"Bai","year":"2021","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chen et al., 2021a] Hao Chen, Benoit Lagadec, and Franc ¸ois Bremond. Ice:inter-instance contrastive encod- ing for unsupervised person re-identification. In ICCV, 2021.","title":"Ice:inter-instance contrastive encoding for unsupervised person re-identification","authors":"Chen","year":"2021","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Chen et al., 2021b] Hao Chen, Yaohui Wang, Benoit La- gadec, Antitza Dantcheva, and Francois Bremond. Joint generative and contrastive learning for unsupervised per- son re-identification. CVPR, 2021.","title":"Joint generative and contrastive learning for unsupervised person re-identification","authors":"Chen","year":"2021","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Deng et al., 2018] Weijian Deng, Liang Zheng, Qixiang Ye, Guoliang Kang, Yi Yang, and Jianbin Jiao. Image- image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification. In CVPR, 2018.","title":"Imageimage domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification","authors":"Weijian Deng, Liang Zheng, Qixiang Ye, Guoliang Kang, Yi Yang, Jianbin Jiao","year":"2018","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Ester et al., 1996] Martin Ester, Hans-Peter Kriegel, Jörg Sander, and Xiaowei Xu. A density-based algorithm for discovering clusters in large spatial databases with noise. In KDD, 1996.","title":"A density-based algorithm for discovering clusters in large spatial databases with noise","authors":"Martin Ester, Hans-Peter Kriegel, Jörg Sander, Xiaowei Xu","year":"1996","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Ge et al., 2020] Yixiao Ge, Feng Zhu, Dapeng Chen, Rui Zhao, and Hongsheng Li. Self-paced contrastive learning with hybrid memory for domain adaptive object re-id. In NeurIPS, 2020.","title":"Self-paced contrastive learning with hybrid memory for domain adaptive object re","authors":"Yixiao Ge, Feng Zhu, Dapeng Chen, Rui Zhao, Hongsheng Li","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Hermans et al., 2017] Alexander Hermans, Lucas Beyer, and Bastian Leibe. In defense of the triplet loss for per- son re-identification. CoRR, abs/1703.07737, 2017.","title":"In defense of the triplet loss for person re-identification","authors":"Alexander Hermans, Lucas Beyer, Bastian Leibe","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Isola et al., 2017] Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A Efros. Image-to-image translation with con- ditional adversarial networks. In CVPR, 2017.","title":"Image-to-image translation with conditional adversarial networks","authors":"Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A Efros","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Jetley et al., 2018] Saumya Jetley, Nicholas A. Lord, Namhoon Lee, and Philip H. S. Torr. Learn to pay at- tention. In ICLR, 2018.","title":"Learn to pay attention","authors":"Saumya Jetley, Nicholas A Lord, Namhoon Lee, Philip H S Torr","year":"2018","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Li and Zhang, 2020] Jianing Li and Shiliang Zhang. Joint visual and temporal consistency for unsupervised domain adaptive person re-identification. In ECCV, 2020.","title":"Joint visual and temporal consistency for unsupervised domain adaptive person re-identification","authors":"Li, Zhang ; ] Jianing Li, Shiliang Zhang","year":"2020","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Lin et al., 2020] Yutian Lin, Lingxi Xie, Yu Wu, Cheng- gang Yan, and Qi Tian. Unsupervised person re- identification via softened similarity learning. In CVPR, 2020.","title":"Unsupervised person reidentification via softened similarity learning","authors":"Yutian Lin, Lingxi Xie, Yu Wu, Chenggang Yan, Qi Tian","year":"2020","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Sun et al., 2018] Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, and Shengjin Wang. Beyond part models: Person retrieval with refined part pooling. In ECCV, 2018.","title":"Beyond part models: Person retrieval with refined part pooling","authors":"Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, Shengjin Wang","year":"2018","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Wang and Zhang, 2020] Dongkai Wang and Shiliang Zhang. Unsupervised person re-identification via multi-label classification. In CVPR, 2020.","title":"Unsupervised person re-identification via multi-label classification","authors":"Wang, Zhang, Dongkai Wang, Shiliang Zhang","year":"2020","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Wang et al., 2018] Guanshuo Wang, Yufeng Yuan, Xiong Chen, Jiwei Li, and Xi Zhou. Learning discrimina- tive features with multiple granularities for person re- identification. In ACM MM, 2018.","title":"Learning discriminative features with multiple granularities for person reidentification","authors":"Wang","year":"2018","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Wang et al., 2020] Zhongdao Wang, Jingwei Zhang, Liang Zheng, Yixuan Liu, Yifan Sun, Yali Li, and Shengjin Wang. Cycas: Self-supervised cycle association for learn- ing re-identifiable descriptions. In ECCV, 2020.","title":"Cycas: Self-supervised cycle association for learning re-identifiable descriptions","authors":"Wang","year":"2020","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Wang et al., 2021] Menglin Wang, Baisheng Lai, Jianqiang Huang, Xiaojin Gong, and Xian-Sheng Hua. Camera- aware proxies for unsupervised person re-identification. In AAAI, 2021.","title":"Cameraaware proxies for unsupervised person re-identification","authors":"Wang","year":"2021","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Wei et al., 2018] Longhui Wei, Shiliang Zhang, Wen Gao, and Qi Tian. Person transfer gan to bridge domain gap for person re-identification. In CVPR, 2018.","title":"Person transfer gan to bridge domain gap for person re-identification","authors":"Wei","year":"2018","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Xuan and Zhang, 2021] Shiyu Xuan and Shiliang Zhang. Intra-inter camera similarity for unsupervised person re- identification. In CVPR, 2021.","title":"Intra-inter camera similarity for unsupervised person reidentification","authors":"Shiyu Xuan, Shiliang Zhang","year":"2021","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Yu et al., 2018] Rui Yu, Zhiyong Dou, Song Bai, Zhaoxiang Zhang, Yongchao Xu, and Xiang Bai. Hard-aware point- to-set deep metric for person re-identification. In ECCV, 2018.","title":"Hard-aware pointto-set deep metric for person re-identification","authors":"Rui Yu, Zhiyong Dou, Song Bai, Zhaoxiang Zhang, Yongchao Xu, Xiang Bai","year":"2018","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Zeng et al., 2020] Kaiwei Zeng, Munan Ning, Yaohua Wang, and Yang Guo. Hierarchical clustering with hard- batch triplet loss for person re-identification. In CVPR. IEEE, 2020.","title":"Hierarchical clustering with hardbatch triplet loss for person re-identification","authors":"Zeng","year":"2020","venue":"CVPR. IEEE","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Zhai et al., 2020] Yunpeng Zhai, Shijian Lu, Qixiang Ye, Xuebo Shan, Jie Chen, Rongrong Ji, and Yonghong Tian. Ad-cluster: Augmented discriminative clustering for do- main adaptive person re-identification. In CVPR, 2020.","title":"Ad-cluster: Augmented discriminative clustering for domain adaptive person re-identification","authors":"Yunpeng Zhai, Shijian Lu, Qixiang Ye, Xuebo Shan, Jie Chen, Rongrong Ji, Yonghong Tian","year":"2020","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhang et al., 2019] Xinyu Zhang, Jiewei Cao, Chunhua Shen, and Mingyu You. Self-training with progressive augmentation for unsupervised cross-domain person re- identification. In ICCV, 2019.","title":"Self-training with progressive augmentation for unsupervised cross-domain person reidentification","authors":"Xinyu Zhang, Jiewei Cao, Chunhua Shen, Mingyu You","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhang et al., 2022] Xinyu Zhang, Dongdong Li, Zhigang Wang, Jian Wang, Errui Ding, Javen Qinfeng Shi, Zhaoxi- ang Zhang, and Jingdong Wang. Implicit sample extension for unsupervised person re-identification. arXiv preprint arXiv:2204.06892, 2022.","title":"Implicit sample extension for unsupervised person re-identification","authors":"Xinyu Zhang, Dongdong Li, Zhigang Wang, Jian Wang, Errui Ding, Javen Qinfeng Shi, Zhaoxiang Zhang, Jingdong Wang","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zheng et al., 2015] Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jingdong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In ICCV, 2015.","title":"Scalable person re-identification: A benchmark","authors":"Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jingdong Wang, Qi Tian","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zheng et al., 2017] Zhedong Zheng, Liang Zheng, and Yi Yang. Unlabeled samples generated by gan improve the person re-identification baseline in vitro. In ICCV, 2017.","title":"Unlabeled samples generated by gan improve the person re-identification baseline in vitro","authors":"Zheng","year":"2017","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zheng et al., 2019] Wenzhao Zheng, Zhaodong Chen, Ji- wen Lu, and Jie Zhou. Hardness-aware deep metric learn- ing. In CVPR, 2019.","title":"Hardness-aware deep metric learning","authors":"Zheng","year":"2019","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhong et al., 2018] Zhun Zhong, Liang Zheng, Zhedong Zheng, Shaozi Li, and Yi Yang. Camera style adaptation for person re-identification. In CVPR, 2018.","title":"Camera style adaptation for person re-identification","authors":"Zhun Zhong, Liang Zheng, Zhedong Zheng, Shaozi Li, Yi Yang","year":"2018","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhu et al., 2017] Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A Efros. Unpaired image-to-image translation using cycle-consistent adversarial networkss. In ICCV, 2017.","title":"Unpaired image-to-image translation using cycle-consistent adversarial networkss","authors":"Zhu","year":"2017","venue":"ICCV","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}