2.793

                    2018影響因子

                    (CJCR)

                    • 中文核心
                    • EI
                    • 中國科技核心
                    • Scopus
                    • CSCD
                    • 英國科學文摘

                    留言板

                    尊敬的讀者、作者、審稿人, 關于本刊的投稿、審稿、編輯和出版的任何問題, 您可以本頁添加留言。我們將盡快給您答復。謝謝您的支持!

                    姓名
                    郵箱
                    手機號碼
                    標題
                    留言內容
                    驗證碼

                    基于圖像和特征聯合約束的跨模態行人重識別

                    張玉康 譚磊 陳靚影

                    張玉康,  譚磊,  陳靚影.  基于圖像和特征聯合約束的跨模態行人重識別.  自動化學報,  2021,  47(8): 1943?1950 doi: 10.16383/j.aas.c200184
                    引用本文: 張玉康,  譚磊,  陳靚影.  基于圖像和特征聯合約束的跨模態行人重識別.  自動化學報,  2021,  47(8): 1943?1950 doi: 10.16383/j.aas.c200184
                    Zhang Yu-Kang,  Tan Lei,  Chen Jing-Ying.  Cross-modality person re-identification based on joint constraints of image and feature.  Acta Automatica Sinica,  2021,  47(8): 1943?1950 doi: 10.16383/j.aas.c200184
                    Citation: Zhang Yu-Kang,  Tan Lei,  Chen Jing-Ying.  Cross-modality person re-identification based on joint constraints of image and feature.  Acta Automatica Sinica,  2021,  47(8): 1943?1950 doi: 10.16383/j.aas.c200184

                    基于圖像和特征聯合約束的跨模態行人重識別

                    doi: 10.16383/j.aas.c200184
                    基金項目: 國家自然科學基金面上項目(61977027), 湖北省科技創新重大專項(2019AAA044)資助
                    詳細信息
                      作者簡介:

                      張玉康:華中師范大學國家數字化學習工程技術研究中心碩士研究生. 主要研究方向為行人重識別, 生成對抗網絡. E-mail: zhangyk@mails.ccnu.edu.cn

                      譚磊:華中師范大學國家數字化學習工程技術研究中心碩士研究生. 主要研究方向為模式識別和計算機視覺. E-mail: lei.tan@mails.ccnu.edu.cn

                      陳靚影:華中師范大學國家數字化學習工程技術研究中心教授. 2001 年獲得南洋理工計算機科學與工程系博士學位. 主要研究方向為圖像處理, 計算機視覺, 模式識別, 多媒體應用. 本文通信作者. E-mail: chenjy@mail.ccnu.edu.cn

                    Cross-modality Person Re-identification Based on Joint Constraints of Image and Feature

                    Funds: Supported by General Program of National Natural Science Foundation of China (61977027), Major scientific and Technological Innovation Projects in Hubei Province (2019AAA044)
                    More Information
                      Author Bio:

                      ZHANG Yu-Kang Master student at the National Engineering Research Center for E-Learning, Central China Normal University. His research interest covers person re-identification and generative adversarial networks

                      TAN Lei Master student at the National Engineering Research Center for E-Learning, Central China Normal University. His research interest covers pattern recognition and computer vision

                      CHEN Jing-Ying Professor at the National Engineering Research Center for E-Learning, Central China Normal University. She received her Ph. D. degree from the School of Computer Engineering, Nanyang Technological University, Singapore in 2001. Her research interest covers image processing, computer vision, pattern recognition, and multimedia applications. Corresponding author of this paper

                    • 摘要:

                      近年來, 基于可見光與近紅外的行人重識別研究受到業界人士的廣泛關注. 現有方法主要是利用二者之間的相互轉換以減小模態間的差異. 但由于可見光圖像和近紅外圖像之間的數據具有獨立且分布不同的特點, 導致其相互轉換的圖像與真實圖像之間存在數據差異. 因此, 本文提出了一個基于圖像層和特征層聯合約束的可見光與近紅外相互轉換的中間模態, 不僅實現了行人身份的一致性, 而且減少了模態間轉換的差異性. 此外, 考慮到跨模態行人重識別數據集的稀缺性, 本文還構建了一個跨模態的行人重識別數據集, 并通過大量的實驗證明了文章所提方法的有效性, 本文所提出的方法在經典公共數據集SYSU-MM01上比D2RL算法在 Rank-1和mAP上分別高出4.2 %和3.7 %, 該方法在本文構建的Parking-01數據集的近紅外檢索可見光模式下比ResNet-50算法在Rank-1和mAP上分別高出10.4 %和10.4 %.

                    • 圖  1  本文方法的總體框架

                      Fig.  1  The overall framework of this method

                      圖  2  數據集圖像示例

                      Fig.  2  Example of dataset images

                      圖  3  中間模態生成器所生成的中間模態圖像

                      Fig.  3  Middle modality image generated by middle modality generator

                      表  1  SYSU-MM01數據集all-search single-shot模式實驗結果

                      Table  1  Experimental results in all-search single-shot mode on SYSU-MM01 dataset

                      方法All-Search Single-shot
                      R1R10R20mAP
                      HOG[19]2.818.332.04.2
                      LOMO[20]3.623.237.34.5
                      Two-Stream[9]11.748.065.512.9
                      One-Stream[9]12.149.766.813.7
                      Zero-Padding[9]14.852.271.416.0
                      BCTR[10]16.254.971.519.2
                      BDTR[10]17.155.572.019.7
                      D-HSME[21]20.762.878.023.2
                      MSR[22]23.251.261.722.5
                      ResNet-50*28.164.677.428.6
                      cmGAN[12]27.067.580.627.8
                      CMGN[23]27.268.281.827.9
                      D2RL[13]28.970.682.429.2
                      本文方法33.173.983.732.9
                      下載: 導出CSV

                      表  2  SYSU-MM01數據集all-search multi-shot模式實驗結果

                      Table  2  Experimental results in all-search multi-shot mode on SYSU-MM01 dataset

                      方法All-Search Multi-shot
                      R1R10R20mAP
                      HOG[19]3.822.837.72.16
                      LOMO[20]4.7028.343.12.28
                      Two-Stream[9]16.458.474.58.03
                      One-Stream[9]16.358.275.18.59
                      Zero-Padding[9]19.261.478.510.9
                      ResNet-50*30.066.275.724.6
                      cmGAN[12]31.572.785.022.3
                      本文方法33.470.078.727.0
                      下載: 導出CSV

                      表  3  SYSU-MM01數據集indoor-search single-shot模式實驗結果

                      Table  3  Experimental results in indoor-search single-shot mode on SYSU-MM01 dataset

                      方法indoor-search single-shot
                      R1R10R20mAP
                      HOG[19]3.224.744.67.25
                      LOMO[20]5.834.454.910.2
                      Two-Stream[9]15.661.281.121.2
                      One-Stream[9]17.063.682.123.0
                      Zero-Padding[9]20.668.485.827.0
                      CMGN[23]30.474.287.540.6
                      ResNet-50*31.078.290.341.9
                      cmGAN[12]31.777.289.242.2
                      本文方法31.179.589.141.3
                      下載: 導出CSV

                      表  4  SYSU-MM01數據集indoor-search multi-shot模式實驗結果

                      Table  4  Experimental results in indoor-search multi-shot mode on SYSU-MM01 dataset

                      方法indoor-search multi-shot
                      R1R10R20mAP
                      HOG[19]4.829.149.43.51
                      LOMO[20]7.440.460.45.64
                      Two-Stream[9]22.572.388.714.0
                      One-Stream[9]22.771.887.915.1
                      Zero-Padding[9]24.575.991.418.7
                      ResNet-50*29.966.275.724.5
                      cmGAN[12]37.080.992.332.8
                      本文方法37.276.083.833.8
                      下載: 導出CSV

                      表  5  近紅外檢索可見光模式的實驗結果

                      Table  5  Experimental results of near infrared retrieval visible mode

                      方法近紅外?>可見光
                      R1R10R20mAP
                      ResNet-50*15.539.751.919.3
                      本文方法25.953.862.829.7
                      下載: 導出CSV

                      表  6  可見光檢索近紅外模式的實驗結果

                      Table  6  Experimental results of visible retrieval near infrared mode

                      方法可見光?>近紅外
                      R1R10R20mAP
                      ResNet-50*20.245.650.014.7
                      本文方法31.648.256.119.7
                      下載: 導出CSV

                      表  7  不同模態轉換的實驗結果

                      Table  7  Experimental results of different mode conversion

                      方法R1R10R20mAP
                      ResNet-50*28.164.677.428.6
                      轉為可見光29.669.880.530.7
                      轉為近紅外30.871.583.231.2
                      本文提出的方法33.173.983.732.9
                      下載: 導出CSV

                      表  8  有無循環一致性損失的實驗結果

                      Table  8  Experimental results with or without loss of cycle consistency

                      方法R1R10R20mAP
                      無循環一致性29.667.178.331.1
                      有循環一致性33.173.983.732.9
                      下載: 導出CSV
                      360彩票
                    • [1] 葉鈺, 王正, 梁超, 韓鎮, 陳軍, 胡瑞敏. 多源數據行人重識別研究綜述. 自動化學報, 2020, 46(9): 1869-1884.

                      Ye Yu, Wang Zheng, Liang Chao, Han Zhen, Chen Jun, Hu Rui-Min. A survey on multi-source person re-identification. Acta Automatica Sinica, 2020, 46(9): 1869?1884
                      [2] 羅浩, 姜偉, 范星, 張思朋. 基于深度學習的行人重識別研究進展[J]. 自動化學報, 2019, 45(11): 2032-2049.

                      LUO Hao, JIANG Wei, FAN Xing, ZHANG Si-Peng. A Survey on Deep Learning Based Person Re-identification. ACTA AUTOMATICA SINICA, 2019, 45(11): 2032-2049.
                      [3] 周勇, 王瀚正, 趙佳琦, 陳瑩, 姚睿, 陳思霖. 基于可解釋注意力部件模型的行人重識別方法. 自動化學報, 2020. doi: 10.16383/j.aas.c200493

                      Zhou Yong, Wang Han-Zheng, Zhao Jia-Qi, Chen Ying, Yao Rui, Chen Si-Lin. Interpretable attention part model for person re-identification. Acta Automatica Sinica, 2020, 41(x): 1?13 doi: 10.16383/j.aas.c200493
                      [4] 李幼蛟, 卓力, 張菁, 李嘉鋒, 張輝. 行人再識別技術綜述[J]. 自動化學報, 2018, 44(9): 1554-1568.

                      LI You-Jiao, ZHUO Li, ZHANG Jing, LI Jia-Feng A Survey of Person Re-identification. ACTA AUTOMATICA SINICA, 2018, 44(9): 1554-1568
                      [5] Zhao H, Tian M, Sun S, et al. Spindle net: Person re-identification with human body region guided feature decomposition and fusion. In: Proceedings of the IEEE CVPR. Hawaii, USA: IEEE, 2017. 1077?1085
                      [6] Sun Y, Zheng L, Yang Y, et al. Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline). In: Proceedings of the ECCV. Munich, Germany: Springer, 2018. 480?496
                      [7] Hermans A, Beyer L, Leibe B. In defense of the triplet loss for person re-identification. arXiv preprint arXiv: 1703.07737, 2017
                      [8] Wei L, Zhang S, Gao W, et al. Person transfer gan to bridge domain gap for person re-identification. In: Proceedings of the IEEE CVPR. Salt Lake City, UT, USA: IEEE, 2018. 79?88
                      [9] Wu A, Zheng W S, Yu H X, et al. RGB-infrared cross-modality person re-identification. In: Proceedings of the IEEE ICCV. Honolulu, USA: IEEE, 2017. 5380?5389
                      [10] Ye M, Wang Z, Lan X, et al. visible Thermal person re-identification via dual-constrained top-ranking. In: Proceeding of IJCAI. Stockholm, Sweden, 2018, 1: 2
                      [11] Ye M, Lan X, Li J, et al. Hierarchical discriminative learning for visible thermal person re-identification. In: Proceeding of AAAI. Louisiana, USA: IEEE, 2018. 32(1)
                      [12] Dai P, Ji R, Wang H, et al. Cross-Modality person re-identification with generative adversarial training. In: Proceeding of IJCAI. Stockholm, Sweden, 2018. 1: 2
                      [13] Wang Z, Wang Z, Zheng Y, et al. Learning to reduce dual-level discrepancy for infrared-visible person re-identification. In: Proceedings of the IEEE CVPR. California, USA: IEEE, 2019. 618?626
                      [14] Wang G, Zhang T, Cheng J, et al. Rgb-infrared cross-modality person re-identification via joint pixel and feature alignment. In: Proceedings of the IEEE ICCV. Seoul, Korea: IEEE, 2019. 3623?3632
                      [15] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE CVPR. Las Vegas, USA: IEEE, 2016. 770?778
                      [16] Zhu J Y, Park T, Isola P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks. In: Proceedings of the IEEE ICCV. Honolulu, USA: IEEE, 2017. 2223?2232
                      [17] Huang R, Zhang S, Li T, et al. Beyond face rotation: Global and local perception gan for photorealistic and identity preserving frontal view synthesis. In: Proceedings of the IEEE ICCV. Honolulu, USA: IEEE, 2017. 2439?2448
                      [18] Deng J, Dong W, Socher R, et al. Imagenet: A large-scale hierarchical image database. In: Proceedings of the IEEE CVPR, Miami, FL, USA: IEEE, 2009. 248?255
                      [19] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: Proceedings of the IEEE CVPR. San Diego, CA, USA: IEEE, 2005: 886?893
                      [20] Liao S, Hu Y, Zhu X, et al. Person re-identification by local maximal occurrence representation and metric learning. In: Proceedings of the IEEE CVPR. Boston, USA: IEEE, 2015: 2197?2206
                      [21] Hao Y, Wang N, Li J, et al. HSME: Hypersphere manifold embedding for visible thermal person re-identification. In: Proceedings of the AAAI. Hawaii, USA: IEEE, 2019, 33: 8385?392
                      [22] Kang J K, Hoang T M, Park K R. Person Re-Identification between Visible and Thermal Camera Images Based on Deep Residual CNN Using Single Input[J]. IEEE Access, 2019: 1-1.
                      [23] B J J A, B K J, B M Q A, et al. A Cross-Modal Multi-granularity Attention Network for RGB-IR Person Re-identification[J]. Neurocomputing, 2020.
                    • 加載中
                    圖(3) / 表(8)
                    計量
                    • 文章訪問數:  375
                    • HTML全文瀏覽量:  186
                    • PDF下載量:  75
                    • 被引次數: 0
                    出版歷程
                    • 收稿日期:  2020-04-03
                    • 錄用日期:  2020-10-19
                    • 網絡出版日期:  2021-01-19
                    • 刊出日期:  2021-08-20

                    目錄

                      /

                      返回文章
                      返回