Asymmetric multi-stage CNNs for small-scale pedestrian detection

Asymmetric multi-stage CNNs for small-scale pedestrian detection
复制标题

用于小规模行人检测的非对称多级 CNN

DOI:
10.1016/j.neucom.2020.05.019
复制
发表时间:
2020-10
期刊:
影响因子:
6
通讯作者:
Xu Changsheng
Xu Changsheng
中科院分区:
计算机科学2区
文献类型:
--
作者:
Zhang Shan;Yang Xiaoshan;Liu Yanxia;Xu Changsheng

文献摘要

参考文献

相似文献

行人检测的一个关键瓶颈是对图像和视频中对比度低、形状模糊的小范围行人的检测。考虑到行人的体形总是长方形(高度大于宽度),我们提出了一种用于小规模行人检测的非对称多级网络(AMS-NET)。提出的方法有两个主要优点。(1)在行人检测中考虑了行人身体形状的不对称性。矩形定位用于生成高度大于宽度的各种矩形方案。此外,采用非对称矩形卷积核函数来捕捉行人身体的紧凑特征。(2)提出的AMS-Net在一个三阶段的框架中,按照从粗到细的特点逐步拒绝非行人包厢。提出的AMS-Net在CATECH测试集的Far子集上显著提高了行人检测的性能(漏检率从60.79%下降到51.36%)。它还在INRIA、ETH、Kitti和CityPeople基准上取得了具有竞争力的表现。
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 features 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.
DOI: 10.1109/iccv.2019.00965
发表时间: 2019-10
期刊: 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
影响因子: --
作者:
Chunluan Zhou;Ming Yang;Junsong Yuan
通讯作者: Chunluan Zhou;Ming Yang;Junsong Yuan
DOI: 10.1007/978-3-030-01246-5_9
发表时间: 2018-09
期刊: --
影响因子: --
作者:
Chunluan Zhou;Junsong Yuan
通讯作者: Chunluan Zhou;Junsong Yuan
DOI: 10.1007/978-1-349-95810-8_460
发表时间: 2018-11
期刊: The Grants Register 2019
影响因子: --
作者:
Die Schulleitung
通讯作者: Die Schulleitung
DOI: 10.1080/01431160412331269698
发表时间: 2005-01-10
影响因子: 3.4
作者:
Pal, M
通讯作者: Pal, M
DOI: 10.1007/s11263-005-6644-8
发表时间: 2005-07-01
影响因子: 19.5
作者:
Viola, P;Jones, MJ;Snow, D
通讯作者: Snow, D