动态查询感知的行人重识别算法

doi:10.3778/j.issn.1002-8331.2306-0312

摘要/Abstract

摘要： 目前无监督的泛化行人重识别算法在某些需要背景信息辅助判断的情况下，可能会忽视图像的局部区域对细粒度特征的关注，导致背景信息被过滤掉，从而降低识别精度。针对上述问题，提出了一种基于稀疏注意力的动态查询感知算法。通过挤压拼接（squeeze and concat，SPC）模块，获取不同通道数的特征图。利用双层路由感知注意力机制，提取不同尺度特征图之间的注意力权重，得到逐级通道注意力向量。对逐级通道注意力向量的权重进行重新校准。将重新标定的权重与相应的特征图进行加权，输出具有更丰富细化特征信息的多尺度特征图。所提模型在大规模公开数据集（Market-1501、DukeMTMC-reID、MSMT17）上进行实验，相较于基线模型Rank-1分别提高了3.2、4.4、15.4个百分点，mAP分别提高了5.5、8.3、16.2个百分点，与现有前沿算法相比，能够实现更好的局部和全局特征通道之间的信息交互，提升模型对图像特征的细节感知能力。

关键词: 行人重识别, 细粒度特征, 稀疏注意力机制, 动态查询感知, 特征重组

Abstract: Currently, unsupervised generalization-based person re-identification algorithms may overlook the fine-grained feature attention in certain situations where background information is needed to assist in judgment, leading to the filtration of background information and a decrease in recognition accuracy. To address this issue, a dynamic query-aware algorithm based on sparse attention is proposed. Firstly, the squeeze and concat (SPC) module is utilized to obtain feature maps with different channel numbers. Secondly, a bi-level routing attention mechanism is employed to extract attention weights between feature maps of different scales, resulting in hierarchical channel attention vectors. Then, the weights of the hierarchical channel attention vectors are recalibrated. Finally, the recalibrated weights are used to weight the corresponding feature maps, thereby generating multi-scale feature maps with richer and more refined feature information. Experimental results on large-scale public datasets (Market-1501, DukeMTMC-reID, MSMT17) demonstrate that the proposed model achieves improvements of 3.2, 4.4, and 15.4 percentage points in Rank-1 accuracy and 5.5, 8.3, and 16.2 percentage points in mAP, respectively. In comparison with existing state-of-the-art algorithms, this model enables better interaction between local and global feature channels, enhancing the model’s ability to perceive detailed image features.

Key words: person re-identification, fine-grained features, sparse attention mechanism, dynamic query-aware, feature recombination

闵锋, 刘煜晖, 毛一新, 况永刚, 刘彪. 动态查询感知的行人重识别算法[J]. 计算机工程与应用, 2024, 60(19): 199-208.

MIN Feng, LIU Yuhui, MAO Yixin, KUANG Yonggang, LIU Biao. Dynamic Query-Aware Person Re-Identification Algorithm[J]. Computer Engineering and Applications, 2024, 60(19): 199-208.

参考文献

[1] ZHU H, KE W, LI D, et al. Dual cross-attention learning for fine-grained visual categorization and object re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2022: 4692-4702.
[2] 厍向阳, 李蕊心, 叶鸥. 融合随机擦除和残差注意力网络的行人重识别[J]. 计算机工程与应用, 2022, 58(3): 215-221.
SHE X Y, LI R X, YE O. Pedestrian re-identification combining random erasing and residual attention network[J]. Computer Engineering and Applications, 2022, 58(3): 215-221.
[3] ZHANG X, LI D, WANG Z, et al. Implicit sample extension for unsupervised person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2022: 7369-7378.
[4] LIAO S, SHAO L. Graph sampling based deep metric learning for generalizable person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition， 2022: 7359-7368.
[5] CHO Y, KIM W J, HONG S, et al. Part-based pseudo label refinement for unsupervised person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 7308-7318.
[6] GU X, CHANG H, MA B, et al. Clothes-changing person re-identification with RGB modality only[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 1060-1069.
[7] 王黎明,孙俊,陈祺东.加强重识别的行人多目标跟踪算法[J].计算机工程与应用, 2022, 58(21): 213-222.
WANG L M, SUN J, CHEN Q D. Pedestrian multi-object tracking algorithm with strengthened re-identification[J].Computer Engineering and Applications, 2022, 58(21): 213-222.
[8] 王仕宸, 黄凯, 陈志刚. 等. 深度学习的三维人体姿态估计综述[J]. 计算机科学与探索, 2023, 17(1): 74-87.
WANG S C, HUANG K, CHEN Z G, et al. Survey on 3D human pose estimation of deep learning[J]. Journal of? Frontiers of Computer Science and Technology, 2023, 17(1): 74-87.
[9] 何坚, 郭泽龙, 刘乐园, 等. 基于滑动窗口和卷积神经网络的可穿戴人体活动识别技术[J]. 电子与信息学报, 2022, 44(1): 168-177.
HE J, GUO Z L, LIU L Y, et al. Human activity recognition technology based on sliding window and convolutional neural network[J]. Electronics & Information Technology, 2022, 44(1): 168-177.
[10] ZHENG L, YANG Y, HAUPTMANN A G. Person re-identification: past, present and future[EB/OL]. [2022-03-15].https://arxiv.org/pdf/1610.02984.pdf.
[11] WANG H, SHEN J, LIU Y, et al. Nformer: robust person re-identification with neighbor transformer[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 7297-7307.
[12] SUN Y F, ZHENG L, YANG Y, et al. Beyond part models: person retrieval with refined part pooling (and a strong convolutional baseline)[C]//Proceedings of the European Conference on Computer Vision, 2018: 501-518.
[13] LIAO S, SHAO L. Interpretable and generalizable person re-identification with query-adaptive convolution and temporal lifting[C]//Proceedings of the16th European Conference on Computer Vision, Glasgow, UK, Aug 23-28, 2020. Cham: Springer International Publishing, 2020: 456-474.
[14] ZHENG Z, YANG X, YU Z, et al. Joint discriminative and generative learning for person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 2138-2147.
[15] 张正一, 丁建伟, 魏慧雯, 等. 基于注意力机制的多级特征级联的行人重识别方法[J]. 激光与光电子学进展, 2021, 58(22): 374-383.
ZHANG Z Y, DING J W, WEI H W, et al. Cascaded multi-level features learning for attention based person re-identification[J]. Laser & Optoelectronics Progress, 2021, 58(22): 374-383.
[16] 李杰. 结合注意力和纹理特征增强的行人再识别[J]. 计算机科学与探索, 2022, 16(3): 661-668.
LI J. Attention and texture feature enhancement for person re-identification[J]. Journal of? Frontiers of Computer Science and Technology, 2022, 16(3): 661-668.
[17] 温静, 张福康. 基于多粒度信息融合的无监督行人重识别方法[J]. 计算机工程与应用, 2023, 59(13): 99-109.
WEN J, ZHANG F K. Unsupervised person re-identification method based on multi-granularity information fusion[J]. Computer Engineering and Applications, 2023, 59(13): 99-109.
[18] 曾涛, 薛峰, 杨添. 面向行人重识别的通道与空间双重注意力网络[J]. 计算机工程, 2022, 48(12): 281-287.
ZENG T, XUE F, YANG T. Channel and spatial dual-attention network for person re-identification[J]. Computer Engineering， 2022, 48(12): 281-287.
[19] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[20] ZHANG H, ZU K, LU J, et al. EPSANet: an efficient pyramid squeeze attention block on convolutional neural network[C]//Proceedings of the Asian Conference on Computer Vision, 2022: 1161-1177.
[21] SUN Y, ZHENG L, DENG W, et al. SVDNet for pedestrian retrieval[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017:3820-3828.
[22] DENG J, GUO J, XUE N, et al. Arcface: additive angular margin loss for deep face recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 4690-4699.
[23] ZHU L, WANG X, KE Z, et al. BiFormer: vision transformer with bi-level routing attention[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 10323-10333.
[24] HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[EB/OL]. [2022-12-13]. https://arxiv.org/pdf/1703.07737.pdf.
[25] IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 2015 International Conference on Machine Learning. New York: ACM, 2015: 448-456.
[26] ZHENG L, SHEN L, TIAN L, et al. Scalable person re-identification: a benchmark[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision, 2016.
[27] HE T, JIN X, SHEN X, et al. Dense interaction learning for video-based person re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 1490-1501.
[28] WEI L, ZHANG S, WEN G, et al. Person transfer GAN to bridge domain gap for person re-identification[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
[29] LUO H, GU Y Z, LIAO X Y, et al. Bag of tricks and a strong baseline for deep person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Long Beach, CA, USA: IEEE Press, 2019: 1487-1495.
[30] LI M, ZHU X, GONG S. Unsupervised tracklet re-identification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 42(7): 1770-1782.
[31] ZHONG Z, ZHENG L, LUO Z, et al. Invariance matters: exemplar memory for domain adaptive re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 598-607.
[32] WANG G, YUAN Y, CHEN X, et al. Learning discriminative features with multiple granularities for person re-identification[C]//Proceedings of the 26th ACM International Conference on Multimedia. New York: ACM, 2018: 274-282.
[33] WANG Z, ZHANG J, ZHENG L, et al. CycAs: self-supervised cycle association for learning re-identifiable descripttions[C]//Proceedings of the European Conference on Computer Vision. Cham: Springer, 2020: 72-88.
[34] GE Y, CHEN D, LI H. Mutual mean-teaching: pseudo label refinery for unsupervised domain adaptation on re-identification[EB/OL]. [2022-03-15]. https://arxiv.org/pdf/2001.01526.pdf.
[35] LI W, ZHU X T, GONG S G, et al. Harmonious attention network for person re-identification[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 2285-2294.
[36] 孙明浩, 王洪元, 吴琳钰, 等. 基于特征金字塔分支和非局部关注的行人重识别[J]. 数据采集与处理, 2023, 38(1): 121-131.
SUN M H, WANG H Y, WU L Y, et al. Person re-identification based on feature pyramid branch and non-local attention[J]. Journal of Data Acquisition and Processing, 2023, 38(1): 121-131.
[37] 朱利, 林欣, 徐亦飞, 等. 基于城市信息单元和差异注意力的多层行人重识别技术[J]. 集成技术, 2023, 12(1): 91-104.
ZHU L, LIN X, XU Y F, et al. Multi-level person re-identification based on urban information unit and diff attention scheme[J]. Journal of Integration Technology, 2023, 12(1): 91-104.
[38] 钱亚萍, 王凤随, 熊磊, 等. 联合特征细化和耐噪声对比学习的无监督行人重识别[J]. 光电子· 激光, 2023, 34(7): 762-770.
QIAN Y P, WANG F S, XIONG L, et al. Joint feature refinement and noise-tolerant comparative learning for unsupervised person re-identification[J]. Journal of Optoelectronics·Laser, 2023, 34(7): 762-770.