文献类型：期刊文献

英文题名：Asymmetric multi-stage CNNs for small-scale pedestrian detection

作者：Zhang, Shan[1];Yang, Xiaoshan[2,3];Liu, Yanxia[4];Xu, Changsheng[2,3]

通讯作者：Liu, YX[1]

机构：[1]Beijing Union Univ, Beijing Key Lab Informat Serv Engn, Beijing 100183, Peoples R China;[2]Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100183, Peoples R China;[3]Univ Chinese Acad Sci, Beijing 100183, Peoples R China;[4]Beijing Union Univ, Coll Urban Rail Transit & Logist, Beijing 100183, Peoples R China

第一机构：北京联合大学北京市信息服务工程重点实验室

通讯机构：[1]corresponding author), Beijing Union Univ, Coll Urban Rail Transit & Logist, Beijing 100183, Peoples R China.|[11417]北京联合大学;

年份：2020

卷号：409

起止页码：12-26

外文期刊名：NEUROCOMPUTING

收录：;EI(收录号：20202408813355);Scopus(收录号：2-s2.0-85086070398);WOS:【SCI-EXPANDED(收录号:WOS:000562543100002)】；

基金：This work was supported by National Key Research and Development Program of China (No. 2018AAA0100604, 2017YFB1002804), National Natural Science Foundation of China (No. 61702511, 61720106006, 61728210, 61751211, 61620106003, 61532009, 61572498, 61572296, 61432019, 61872424, 61602041, U1836220, U1705262) and Key Research Program of Frontier Sciences of CAS (QYZDJSSWJSC039). This work was also supported by Research Program of National Laboratory of Pattern Recognition (No. Z-2018007) and the Premium Funding Project for Academic Human Resources Development in Beijing Union University (NO. BPHR2020BZ02).

语种：英文

外文关键词：Pedestrian detection; Deep learning; Convolution kernels

摘要：A critical bottleneck in pedestrian detection is the detection of small-scale pedestrians, which have low contrast and blurry shapes in images and videos. Considered that the body shape of a pedestrian is always rectangular (the height is greater than the width), we propose an asymmetric multi-stage network (AMS-Net) for small-scale pedestrian detection. The proposed method has two main advantages. (1) It considers the asymmetry of a pedestrian's body shape in pedestrian detection. The rectangular anchors are used to generate various rectangular proposals that have a height greater than the width. In addition, asymmetric rectangular convolution kernels are adopted for capturing the compact features of the pedestrian body. (2) The proposed AMS-Net gradually rejects the non-pedestrian boxes according to coarse-to-fine fea-tures in a three-stage framework. The proposed AMS-Net significantly improves the performance of pedestrian detection on the Far subset of the Caltech testing set (the miss rate decreases from 60.79% to 51.36%). It also achieves competitive performance on the INRIA, ETH, KITTI and CityPersons benchmarks. (C) 2020 Elsevier B.V. All rights reserved.