| 深度可分离卷积和标准卷积相结合的高效行人检测器 |
| |
| 引用本文： | 张运波,易鹏飞,周东生,张强,魏小鹏. 深度可分离卷积和标准卷积相结合的高效行人检测器[J]. 图学学报, 2022, 43(2): 230-238. DOI: 10.11996/JG.j.2095-302X.2022020230 |
| |
| 作者姓名： | 张运波 易鹏飞 周东生 张强 魏小鹏 |
| |
| 作者单位： | 1. 大连大学软件工程学院先进设计与智能计算省部共建教育部重点实验室，辽宁 大连 116622；2. 大连理工大学计算机科学与技术学院，辽宁 大连 116024 |
| |
| 基金项目： | 大连市双重项目;大连市及大连大学创新团队资助计划;辽宁省中央指导地方科技发展专项;辽宁特聘教授资助计划;辽宁省高等学校;国家自然科学基金 |
| |
| 摘 要： | 行人检测器对算法的速度和精确度有很高的要求.虽然基于深度卷积神经网络(DCNN)的行人检测器具有较高的检测精度,但是这类检测器对硬件设备的计算能力要求较高,因此,这类行人检测器无法很好地部署到诸如移动设备、嵌入式设备和自动驾驶系统等轻量化系统中.基于此,提出了一种更好地平衡速度和精度的轻量级行人检测器(EPDNet)....
|
| 关 键 词： | 标准卷积 深度可分离卷积 特征融合 轻量化 行人检测 |
Efficient pedestrian detector combining depthwise separable convolution and standard convolution |
| |
| ZHANG Yun-bo,YI Peng-fei,ZHOU Dong-sheng,ZHANG Qiang,WEI Xiao-peng. Efficient pedestrian detector combining depthwise separable convolution and standard convolution[J]. Journal of Graphics, 2022, 43(2): 230-238. DOI: 10.11996/JG.j.2095-302X.2022020230 |
| |
| Authors: | ZHANG Yun-bo YI Peng-fei ZHOU Dong-sheng ZHANG Qiang WEI Xiao-peng |
| |
| Affiliation: | 1. Key Laboratory of Advanced Design and Intelligent Computing (Dalian University), Ministry of Education, Dalian Liaoning, 116622, China;2. School of Computer Science and Technology, Dalian University of Technology, Dalian Liaoning, 116024, China |
| |
| Abstract: | Pedestrian detectors require the algorithm to be fast and accurate. Although pedestrian detectors based on deepconvolutional neural networks (DCNN) have high detection accuracy, such detectors require higher capacity ofcalculation. Therefore, such pedestrian detectors cannot be deployed well on lightweight systems, such as mobile devices,embedded devices, and autonomous driving systems. Considering these problems, a lightweight and effective pedestrian detector (EPDNet) was proposed, which can better balance speed and accuracy. First, the shallow convolution layers of the backbone network employed depthwise separable convolution to compress the parameters of model, and the deeper convolution layers utilized standard convolution to extract high-level semantic features. In addition, in order to further improve the performance of the model, the backbone network adopted a feature fusion method to enhance the expression ability of its output features. Through comparative experiments, EPDNet has shown superior performance on twochallenging pedestrian datasets, Caltech and CityPersons. Compared with the benchmark model, EPDNet has obtained abetter trade-off between speed and accuracy, improving the speed and accuracy of EPDNet at the same time. |
| |
| Keywords: | standard convolution depthwise separable convolution feature fusion lightweight pedestrian detection |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《图学学报》浏览原始摘要信息 |
|
点击此处可从《图学学报》下载免费的PDF全文 |
|
