Pedestrian Attributes Recognition in Surveillance Scenarios with Hierarchical Multi-Task CNN Models

作     者：Wenhua Fang Jun Chen Ruimin Hu 

作者机构：National Engineering Research Center for Multimedia SoftwareComputer School of Wuhan UniversityWuhan 430072China Hubei Key Laboratory of Multimedia and Network Communication EngineeringWuhan UniversityWuhan 430072China Collaborative Innovation Center of Geospatial TechnologyWuhan 430079China 

出 版 物：《China Communications》 (中国通信（英文版）)

年 卷 期：2018年第15卷第12期

页      面：208-219页

核心收录：

学科分类：0810[工学-信息与通信工程] 0808[工学-电气工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 0839[工学-网络空间安全] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by National Key R&D Program of China(-NO.2017YFC0803700) National Nature Science Foundation of China(No.U1736206) National Nature Science Foundation of China(61671336) National Nature Science Foundation of China(61671332) Technology Research Program of Ministry of Public Security(No.2016JSYJA12) Hubei Province Technological Innovation Major Project(-No.2016AAA015) Hubei Province Technological Innovation Major Projec(2017AAA123) National Key Research and Development Program of China(No.2016YFB0100901) Nature Science Foundation of Jiangsu Province(No.BK20160386) 

主　　题：attributes recognition CNN multi-task learning 

摘      要：Pedestrian attributes recognition is a very important problem in video surveillance and video forensics. Traditional methods assume the pedestrian attributes are independent and design handcraft features for each one. In this paper, we propose a joint hierarchical multi-task learning algorithm to learn the relationships among attributes for better recognizing the pedestrian attributes in still images using convolutional neural networks(CNN). We divide the attributes into local and global ones according to spatial and semantic relations, and then consider learning semantic attributes through a hierarchical multi-task CNN model where each CNN in the first layer will predict each group of such local attributes and CNN in the second layer will predict the global attributes. Our multi-task learning framework allows each CNN model to simultaneously share visual knowledge among different groups of attribute categories. Extensive experiments are conducted on two popular and challenging benchmarks in surveillance scenarios, namely, the PETA and RAP pedestrian attributes datasets. On both benchmarks, our framework achieves superior results over the state-of-theart methods by 88.2% on PETA and 83.25% on RAP, respectively.