07 基于动态前景聚焦与伪孪生网络的跨分辨率行人重识别

吉海瑞;张宝华;

1:内蒙古科技大学信息工程学院

摘要(Abstract)：

针对跨分辨率场景下行人图像存在场景复杂、重建图像特征提取效果差等问题，提出基于动态前景聚焦与伪孪生网络的跨分辨率行人重识别算法.该算法在超分辨率重建网络中嵌入动态前景聚焦模块，利用全卷积自动编码器提取目标行人特征，通过高斯掩码对网络进行空间引导，从而使判别特征聚焦在前景上；并经过动态感知模块自动捕获前景的重要特征.又通过构建多粒度相互协同的伪孪生网络，实现判别特征的精细化识别.最后，所提算法在跨分辨率数据集MLR-Market-1501,MLR-DukeMTMC-ReID和CAVIAR上进行实验，Rank-1精度分别达到了91.3%,83.4%和48.5%,证明了所提算法对跨分辨率行人重识别任务的有效性.

关键词(KeyWords)： 行人重识别;跨分辨率;动态前景聚焦;伪孪生网络

基金项目(Foundation): 国家自然科学基金资助项目(61962046,62262048,62001255,62066036,61841204)

作者(Authors): 吉海瑞;张宝华;

DOI: 10.16559/j.cnki.2095-2295.2023.01.007

参考文献(References)：

• [1] JIAO J,ZHENG W S,WU A,et al.Deep low-resolution person re-identification[C]//Proceedings of the AAAI Conference on Artificial Intelligence.Menlo Park,USA:AAAI,2018:6665.

• [2] WANG Z,YE M,YANG F,et al.Cascaded SR-GAN for scale-adaptive low resolution person re-identification[C]//International Joint Conference on Artificial Intelligence.California,USA:IJCAI,2018:3662.

• [3] CHENG Z,DONG Q,GONG S,et al.Inter-task association critic for cross-resolution person re-identification[C]// Proceedings of the IEEE/CVF conference on computer vision and pattern recognition.Long Beach,CA,USA:IEEE,2020:2605.

• [4] ZHANG G,GE Y,DONG Z,et al.Deep high-resolution representation learning for cross-resolution person re-identification[J].IEEE Transactions on Image Processing,2021,30:8913.

• [5] LIU H,TAN X,ZHOU X.Parameter sharing exploration and hetero-center triplet loss for visible-thermal person re-identification[J].IEEE Transactions on Multimedia,2020,23:4414.

• [6] ZHENG L,SHEN L,TIAN L,et al.Scalable person re-identification:A benchmark[C]//Proceedings of the IEEE international conference on computer vision.Boston,MA,USA:IEEE,2015:1116.

• [7] ZHENG Z,ZHENG L,YANG Y.Unlabeled samples generated by gan improve the person re-identification baseline in vitro[C]//Proceedings of the IEEE international conference on computer vision.Honolulu,HI,USA:IEEE,2017:3754.

• [8] CHENG D S,CRISTANI M,STOPPA M,et al.Custom pictorial structures for re-identification[C]// Proceedings of British Machine Vision Conference.Dundee,UK:BMVC,2011:6.

• [9] ZHOU K,YANG Y,CAVALLARO A,et al.Omni-scale feature learning for person re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Long Beach,CA,USA:IEEE,2019:3702.

• [10] MEI Y,FAN Y,Zhou Y.Image super-resolution with non-local sparse attention[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Nashville,TN,USA:IEEE,2021:3517.

• [11] ZHU X,GUO K,REN S,et al.Lightweight image super-resolution with expectation-maximization attention mechanism[J].IEEE Transactions on Circuits and Systems for Video Technology,2021,32(3):1273.

• [12] MAO S,ZHANG S,YANG M.Resolution-invariant person re-identification[J].arXiv preprint arXiv:1906.09748,2019.

• [13] LI Y J,CHEN Y C,LIN Y Y,et al.Recover and identify:A generative dual model for cross-resolution person re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Long Beach,CA,USA:IEEE,2019:8090.

• [14] LI Y J,CHEN Y C,LIN Y Y,et al.Cross-resolution adversarial dual network for person re-identification and beyond[J].arXiv preprint arXiv:2002.09274,2020.

• [15] MUNIR A,LYU C,Goossens B,et al.Resolution based Feature Distillation for Cross Resolution Person Re-Identification[C]//Proceedings of the IEEE/CVF international conference on computer vision.Nashville,TN,USA:IEEE,2021:281.