数据驱动和基于语义的行人再识别

Fangjie Xu, Keyang Cheng, Kaifa Hui, Jianming Zhang
{"title":"数据驱动和基于语义的行人再识别","authors":"Fangjie Xu, Keyang Cheng, Kaifa Hui, Jianming Zhang","doi":"10.1109/FSKD.2017.8393408","DOIUrl":null,"url":null,"abstract":"Pedestrian Re-identification faces many difficulties in training of supervised model because of limited number of labeled data of surveillance videos. Besides, applications of pedestrian re-identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation. In this paper, a data-driven pedestrian re-identification model based on hierarchical semantic representation is proposed, this model extracting essential features with unsupervised deep learning model and enhancing the semantic representation of features with hierarchical mid-level attributes. Firstly, CNNs, well-trained with the training process of CAEs, is used to extract features of horizontal blocks segmented from unlabeled pedestrian images. Then, these features are input into corresponding attribute classifiers to judge whether the pedestrian has the attributes. Lastly, with a table of ‘attributes-classes mapping relations’, final result can be calculated. Our method is proved to significantly outperform the state of the art on the VIPeR and i-LIDS data set in the aspects of accuracy and semanteme.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"240 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Data-driven and semantic-based pedestrian re-identification\",\"authors\":\"Fangjie Xu, Keyang Cheng, Kaifa Hui, Jianming Zhang\",\"doi\":\"10.1109/FSKD.2017.8393408\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Pedestrian Re-identification faces many difficulties in training of supervised model because of limited number of labeled data of surveillance videos. Besides, applications of pedestrian re-identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation. In this paper, a data-driven pedestrian re-identification model based on hierarchical semantic representation is proposed, this model extracting essential features with unsupervised deep learning model and enhancing the semantic representation of features with hierarchical mid-level attributes. Firstly, CNNs, well-trained with the training process of CAEs, is used to extract features of horizontal blocks segmented from unlabeled pedestrian images. Then, these features are input into corresponding attribute classifiers to judge whether the pedestrian has the attributes. Lastly, with a table of ‘attributes-classes mapping relations’, final result can be calculated. Our method is proved to significantly outperform the state of the art on the VIPeR and i-LIDS data set in the aspects of accuracy and semanteme.\",\"PeriodicalId\":236093,\"journal\":{\"name\":\"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)\",\"volume\":\"240 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-07-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/FSKD.2017.8393408\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/FSKD.2017.8393408","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

由于监控视频的标记数据数量有限,行人再识别在监督模型的训练中遇到了很多困难。此外,由于缺乏语义表征,行人再识别在行人检索和犯罪跟踪中的应用受到限制。本文提出了一种基于分层语义表示的数据驱动行人再识别模型,该模型利用无监督深度学习模型提取基本特征,增强具有分层中层属性特征的语义表示。首先,利用经过cae训练过程训练好的cnn,从未标记的行人图像中提取水平块的特征。然后,将这些特征输入到相应的属性分类器中,判断行人是否具有这些属性。最后,利用“属性-类映射关系”表,计算出最终结果。事实证明,我们的方法在准确性和语义方面明显优于VIPeR和i-LIDS数据集的最新技术。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Data-driven and semantic-based pedestrian re-identification
Pedestrian Re-identification faces many difficulties in training of supervised model because of limited number of labeled data of surveillance videos. Besides, applications of pedestrian re-identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation. In this paper, a data-driven pedestrian re-identification model based on hierarchical semantic representation is proposed, this model extracting essential features with unsupervised deep learning model and enhancing the semantic representation of features with hierarchical mid-level attributes. Firstly, CNNs, well-trained with the training process of CAEs, is used to extract features of horizontal blocks segmented from unlabeled pedestrian images. Then, these features are input into corresponding attribute classifiers to judge whether the pedestrian has the attributes. Lastly, with a table of ‘attributes-classes mapping relations’, final result can be calculated. Our method is proved to significantly outperform the state of the art on the VIPeR and i-LIDS data set in the aspects of accuracy and semanteme.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Space syntax and time distance based analysis on the influences of the subways to the pubic traffic accessibility in Nanchang city Designing fuzzy apparatus to model dyslexic individual symptoms for clinical use A kNN classifier optimized by P systems Research on optimal operation of cascade hydropower station based on improved biogeography-based optimization algorithm An estimation algorithm of time-varying channels in the OFDM communication system
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1