融合Gabor特征与卷积特征的小样本行人重识别

doi:10.6040/j.issn.1672-3961.0.2020.232

摘要/Abstract

摘要： 针对视频监控环境下采集的可用行人图像数量有限,以及非可靠数据标注导致监督学习算法性能下降等问题,提出一种融合Gabor特征和卷积特征的无监督小样本行人重识别方法。采用Gabor变换提取多尺度、多方向行人纹理和边缘信息,实现小样本行人图像特征级数据增强,进一步通过特征编码消除冗余信息,提升相似度比对效率。采用卷积自编码网络提取行人非线性深度卷积特征,避免监督学习算法对数据标注的依赖性。融合两种异构特征用于行人相似度比对,实现小样本下行人特征数据的拓展,同时实现行人特征判别能力增强。在Market-1501和DukeMTMC-reID数据集的试验中rank-1准确度分别达到74%和67.1%,证明所提网络架构能有效提升小样本行人重识别的性能。

关键词: Gabor变换, 特征编码, 卷积自编码网络, 无监督学习, 小样本行人重识别

Abstract: In the video surveillance, the limited available person images and unreliable data annotation led to the performance degradation of supervised person re-identification. To solve these problems, we proposed an unsupervised small sample person re-identification method that integrated Gabor features and convolution features. Gabor transform was used to extract multi-scale and multi-direction person texture and edge information, so as to realize the data augmentation of small sample person images in feature level. The redundant information was eliminated by feature encoding to improve the efficiency of feature similarity calculation. The convolutional auto-encoder network was adopted to extract the nonlinear deep convolution feature of pedestrian, which avoided the dependence of supervised learning algorithm on data annotation. The fusion of two heterogeneous features was applied to person similarity comparison, which implemented the feature augmentation of small samples and the improvement of person feature discrimination ability. Experiments were implemented based on Market-1501 and DukeMTMC-reID datasets, the rank-1 accuracy reached 74% and 67.1% respectively. The experimental results showed that the proposed network framework effectively improved the performance of small sample person re-identification.

Key words: Gabor transform, feature encoding, convolutional auto-encoder network, unsupervised learning, small sample person re-identification

中图分类号:

TP391

傅桂霞,邹国锋,毛帅,潘金凤,尹丽菊. 融合Gabor特征与卷积特征的小样本行人重识别[J]. 山东大学学报 (工学版), 2021, 51(3): 22-29.

FU Guixia, ZOU Guofeng, MAO Shuai, PAN Jinfeng, YIN Liju. Small sample person re-identification combining Gabor features and convolution features[J]. Journal of Shandong University(Engineering Science), 2021, 51(3): 22-29.

参考文献 20

[1]	BEDAGKAR G A, SHAH S K. A survey of approaches and trends in person re-identification[J]. Image and Vision Computing, 2014, 32(4):270-286.
[2]	李幼蛟,卓力,张菁,等. 行人再识别技术综述[J]. 自动化学报, 2018, 44(9): 1554-1568. LI Y J, ZHUO L, ZHANG J, et al. A survey of person re-identification[J]. Acta Automatica Sinica, 2018, 44(9): 1554-1568.
[3]	LV J M, CHEN W H, LI Q, et al. Unsupervised cross-dataset person re-identification by transfer learning of spatial-temporal patterns[C] //Proceeding of the 2018 IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE, 2018: 7948-7956.
[4]	ZHONG Z, ZHENG L, ZHENG Z D, et al. CamStyle: a novel data augmentation method for person re-identi-fication[J]. IEEE Transaction on Image Processing, 2019, 28(3): 1176-1190.
[5]	金翠,王洪元,陈首兵. 基于随机擦除行人对齐网络的行人重识别方法[J]. 山东大学学报(工学版),2018,48(6): 67-73. JIN Cui, WANG Hongyuan, CHEN Shoubing. Person re-identification based on random erasing pedestrian alignment network method[J]. Journal of Shandong University(Engineering Science), 2018, 48(6): 67-73.
[6]	CHEN Y C, ZHU X T, ZHENG W S, et al. Person re-identification by camera correlation aware feature augmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(2): 392-408.
[7]	ZHAO C R, WANG X K, MIAO D Q, et al. Maximal granularity structure and generalized multi-view discri-minant analysis for person re-identification[J]. Pattern Recognition, 2018, 79: 79-96.
[8]	VINYALS O, BLUNDELL C, LILLICRAP T, et al. Matching networks for one shot learning[C] //Proceeding of the 2016 Conference on Neural Information Processing Systems. Barcelona, Spain: NIPS, 2016: 3630-3638.
[9]	LIAO S, HU Y, ZHU X, et al. Person re-identification by local maximal occurrence representation and metric learning[C] //Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA: IEEE, 2015: 2197-2206.
[10]	WANG H, GONG S, XIANG T. Unsupervised learning of generative topic saliency for person re-identification[C] //Proceedings of the 2014 British Machine Vision Conference. Nottingham, UK: Springer, 2014.
[11]	YAN C, LUO M, LIU W, et al. Robust dictionary learning with graph regularization for unsupervised person re-identification[J]. Multimedia Tools and Applications, 2018, 77(3): 3553-3577.
[12]	DAUGMAN J. Uncertainty relation for resolution in space, spatial, frequency, and orientation optimized by two-dimensional visual cortical filters[J]. Journal of the Optical Society of America, 1985, 2(7): 1160-1169.
[13]	RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[M]. Cambridge, Massachusetts, USA: MIT Press, 1986:533-536.
[14]	MASCI J, MEIER U, CIRESAN D, et al. Stacked convolutional auto-encoders for hierarchical feature extraction [C] //Proceedings of the 2011 International Conference on Artificial Neural Networks. Espoo, Finland: Springer, 2011: 52-59.
[15]	ZHENG L, SHEN L Y, TIAN L, et al. Scalable person re-identification: a benchmark[C] //Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago, Chile: IEEE, 2015:1116-1124.
[16]	RISTANI E, SOLERA F, ZOU R, et al. Performance measures and a data set for multi-target, multi camera tracking[C] //Proceedings of the 2016 European Conference on Computer Vision. Cham, Switzerland: Spr-inger, 2016: 17-35.
[17]	LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[18]	NG A. Sparse auto-encoder[R]. Stanford University, USA: CS294A Lecture Notes, 2011, 72: 1-19.
[19]	BENGIO Y, LAMBLIM P, POPPVIVI D, et al. Greedy layer-wise training of deep networks[C] //Proceedings of the 2007 Conference on Neural Information Processing Systems. Vancouver, Canada: NIPS, 2007: 153-160.
[20]	RADFORD A, METZ, CHINTALA S. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks[C] //Proceedings of the 4th International Conference on Learning Representations. San Juan, Puerto Rico: ICLR, 2016: 1-16.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed