基于深度学习与属性学习相结合的行人再识别
|
摘要
现实情况中缺少大量有标签数据,导致有监督的行人再识别模型训练受到影响。此外,低层特征的缺乏语义特性限制了行人再识别在行人检索、罪犯追踪等中的应用。本文提出了一种基于深度学习与属性学习相结合的行人再识别方法,利用深度学习的无监督模型提取行人图像的本质特征,并引入"属性"概念增强特征的语义表达能力。首先采用卷积自动编码器进行无监督的特征提取,提取的特征然后交由多个属性分类器进行属性分类,并结合统计获得的属性类别映射关系表计算最终类别判定,最后在VIPeR和i-LIDS标准数据集上进行了测试,并与基于优化属性的行人再识别方法(Optimized attribute based re-identification,OAR)、显著性检测对应法(Salience detection correspondence,SDC)等进行了比较,结果表明本方法能够赋予行人再识别较好的语义性能,并在一定程度上提高了识别的准确率,同时获得了较好的零训练样本识别效果。
The lack of much labeled data in the real world affects the training of supervised model for pedestrian re-identification.Besides,applications of pedestrian re-identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation.This paper presents a pedestrian re-identification method based on the combination of deep learning and attributes learning,which extracts essential features with unsupervised deep learning model and enhances the semantic representation of features with ‘attributes'.Firstly,a convolutional auto-encoder(CAE)is used to extract features of unlabeled pedestrian images,and the extracted features are then input into several attribute classifiers to judge whether the pedestrian owns the attributes.Lastly,with a table of 'attributes-classes mapping relations',we can get the final dassification result.Tests of the proposed algorithm and comparisons with other algorithms on the VIPeR and i-LIDS datasets are shown,and results prove that our algorithm indeed strengthens the semantic representation and improves the accuracy of pedestrian re-identification,achieving good‘zero-shot're-identification performance as well.
The lack of much labeled data in the real world affects the training of supervised model for pedestrian re-identification.Besides,applications of pedestrian re-identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation.This paper presents a pedestrian re-identification method based on the combination of deep learning and attributes learning,which extracts essential features with unsupervised deep learning model and enhances the semantic representation of features with ‘attributes'.Firstly,a convolutional auto-encoder(CAE)is used to extract features of unlabeled pedestrian images,and the extracted features are then input into several attribute classifiers to judge whether the pedestrian owns the attributes.Lastly,with a table of 'attributes-classes mapping relations',we can get the final dassification result.Tests of the proposed algorithm and comparisons with other algorithms on the VIPeR and i-LIDS datasets are shown,and results prove that our algorithm indeed strengthens the semantic representation and improves the accuracy of pedestrian re-identification,achieving good‘zero-shot're-identification performance as well.
引文
[1]Farenzena M,Bazzani L,Perina A,et al.Person re-identification by symmetry-driven accumulation of local features[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.San Francisco,USA:IEEE,2010:2360-2367.
[2]Freeman W T,Roth M.Orientation histograms for hand gesture recognition[C]//International Work-shop on Automatic Face and Gesture Recognition.Zurich,Swiss:IEEE,1995,12:296-301.
[3]Huo Zhonghua,Chen Ying.Person re-identification based on multi-salience fusion[J].Opto-Electronic Engineering,2015,42(9):41-47.
[4]Zheng W S,Gong S,Xiang T.Re-identification by relative distance comparison[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(3):653-668.
[5]Zhao R,Ouyang W,Wang X.Unsupervised salience learning for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Portland:IEEE,2013:3586-3593.
[6]范彩霞,陈亚军,曹磊,等.基于视觉感知模型的行人再识别[J].计算机工程与应用,2016,52(6):156-161.Fan Caixia,Chen Yajun,Cao Lei,et al.Person re-identification based on visual perceptual model[J].Computer Engineering and Applications,2016,52(6):156-161.
[7]卢宏涛,张秦川.深度卷积神经网络在计算机视觉中的应用研究综述[J].数据采集与处理,2016,31(1):1-17.Lu Hongtao,Zhang Qinchuan.Application of deep convolutional neural network in computer vision[J].Journal of Data Acquisition and Processing,2016,31(1):1-17.
[8]LeCun Y,Bengio Y,Hinton G.Deep learning[J].Nature,2015,521(7553):436-444.
[9]曲建岭,杜辰飞,邸亚洲,等.深度自动编码器的研究与展望[J].计算机与现代化,2014(8):128-134.Qu Jiangling,Du Chenfei,Di Yazhou,et al.Research and prospect of deep auto-encoders[J].Computer and Moderni-zation,2014(8):128-134.
[10]Masci J,Meier U,Ciresan D,et al.Stacked convolutional auto-encoders for hierarchical feature extraction[C]//Artificial Neural Networks and Machine Learning-ICANN 2011.Berlin Heidelberg:Springer,2011:52-59.
[11]Gray D,Tao H.Viewpoint invariant pedestrian recognition with an ensemble of localized features[C]//Proceeding of the10th European Conference on Computer Vision.Marseille,France:Springer-Verlag,2008,5302:262-275.
[12]Gray D,Tao H.Viewpoint invariant pedestrian recognition with an ensemble of localized features[C]//Computer Vision-ECCV 2008.Berlin Heidelberg:Springer,2008,5302:262-275.
[13]Layne R,Hospedales T M,Gong S.Attributes-based re-identification[M]//Person Re-identification.London:Springer,2014:93-117.
[14]Umeda T,Sun Y,Irie G,et al.Attribute discovery for person re-identification[C]//International Conference on MultiMedia Modeling.North Miami:Springer International Publishing,2016:268-276.
[15]Ahmed E,Jones M,Marks T K.An improved deep learning architecture for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2015:3908-3916.
[16]Zheng W S,Gong S,Xiang T.Associating groups of people[C]//Proceedings of the 20th British Machine Vision Conference(BMVC).London:IEEE,2009:1-11.
[17]Layne R,Hospedales T M,Gong S,et al.Person re-identification by attributes[C]//Proceedings of the 23rd British Machine Vision Conference(BMVC).Guildford:IEEE,2012,2(3):8-18.
[2]Freeman W T,Roth M.Orientation histograms for hand gesture recognition[C]//International Work-shop on Automatic Face and Gesture Recognition.Zurich,Swiss:IEEE,1995,12:296-301.
[3]Huo Zhonghua,Chen Ying.Person re-identification based on multi-salience fusion[J].Opto-Electronic Engineering,2015,42(9):41-47.
[4]Zheng W S,Gong S,Xiang T.Re-identification by relative distance comparison[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(3):653-668.
[5]Zhao R,Ouyang W,Wang X.Unsupervised salience learning for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Portland:IEEE,2013:3586-3593.
[6]范彩霞,陈亚军,曹磊,等.基于视觉感知模型的行人再识别[J].计算机工程与应用,2016,52(6):156-161.Fan Caixia,Chen Yajun,Cao Lei,et al.Person re-identification based on visual perceptual model[J].Computer Engineering and Applications,2016,52(6):156-161.
[7]卢宏涛,张秦川.深度卷积神经网络在计算机视觉中的应用研究综述[J].数据采集与处理,2016,31(1):1-17.Lu Hongtao,Zhang Qinchuan.Application of deep convolutional neural network in computer vision[J].Journal of Data Acquisition and Processing,2016,31(1):1-17.
[8]LeCun Y,Bengio Y,Hinton G.Deep learning[J].Nature,2015,521(7553):436-444.
[9]曲建岭,杜辰飞,邸亚洲,等.深度自动编码器的研究与展望[J].计算机与现代化,2014(8):128-134.Qu Jiangling,Du Chenfei,Di Yazhou,et al.Research and prospect of deep auto-encoders[J].Computer and Moderni-zation,2014(8):128-134.
[10]Masci J,Meier U,Ciresan D,et al.Stacked convolutional auto-encoders for hierarchical feature extraction[C]//Artificial Neural Networks and Machine Learning-ICANN 2011.Berlin Heidelberg:Springer,2011:52-59.
[11]Gray D,Tao H.Viewpoint invariant pedestrian recognition with an ensemble of localized features[C]//Proceeding of the10th European Conference on Computer Vision.Marseille,France:Springer-Verlag,2008,5302:262-275.
[12]Gray D,Tao H.Viewpoint invariant pedestrian recognition with an ensemble of localized features[C]//Computer Vision-ECCV 2008.Berlin Heidelberg:Springer,2008,5302:262-275.
[13]Layne R,Hospedales T M,Gong S.Attributes-based re-identification[M]//Person Re-identification.London:Springer,2014:93-117.
[14]Umeda T,Sun Y,Irie G,et al.Attribute discovery for person re-identification[C]//International Conference on MultiMedia Modeling.North Miami:Springer International Publishing,2016:268-276.
[15]Ahmed E,Jones M,Marks T K.An improved deep learning architecture for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2015:3908-3916.
[16]Zheng W S,Gong S,Xiang T.Associating groups of people[C]//Proceedings of the 20th British Machine Vision Conference(BMVC).London:IEEE,2009:1-11.
[17]Layne R,Hospedales T M,Gong S,et al.Person re-identification by attributes[C]//Proceedings of the 23rd British Machine Vision Conference(BMVC).Guildford:IEEE,2012,2(3):8-18.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |