Monocular 3d object detection based on uncertainty prediction of keypoints
文献类型:期刊论文
作者 | Chen M(陈牧)1,2,3,4; Zhao HC(赵怀慈)2,3,4; Liu PF(刘鹏飞)1,2,3 |
刊名 | Machines |
出版日期 | 2022 |
卷号 | 10期号:1页码:1-16 |
ISSN号 | 2075-1702 |
关键词 | Keypoints Monocular 3D detection Uncertainty prediction |
产权排序 | 1 |
英文摘要 | Three-dimensional (3D) object detection is an important task in the field of machine vision, in which the detection of 3D objects using monocular vision is even more challenging. We observe that most of the existing monocular methods focus on the design of the feature extraction framework or embedded geometric constraints, but ignore the possible errors in the intermediate process of the detection pipeline. These errors may be further amplified in the subsequent processes. After exploring the existing detection framework of keypoints, we find that the accuracy of keypoints prediction will seriously affect the solution of 3D object position. Therefore, we propose a novel keypoints uncertainty prediction network (KUP-Net) for monocular 3D object detection. In this work, we design an uncertainty prediction module to characterize the uncertainty that exists in keypoint prediction. Then, the uncertainty is used for joint optimization with object position. In addition, we adopt position-encoding to assist the uncertainty prediction, and use a timing coefficient to optimize the learning process. The experiments on our detector are conducted on the KI TTI benchmark. For the two levels of easy and moderate, we achieve accuracy of 17.26 and 11.78 in AP3D, and achieve accuracy of 23.59 and 16.63 in APBEV, which are higher than the latest method KM3D. |
资助项目 | National Equipment Development Department of China[41401040105] ; National Natural Science Foundation of China[U2013210] |
WOS研究方向 | Engineering |
语种 | 英语 |
WOS记录号 | WOS:000747886900001 |
资助机构 | National Equipment Development Department of China (Grant No. 41401040105) ; National Natural Science Foundation of China (Grant No. U2013210) |
源URL | [http://ir.sia.cn/handle/173321/30230] |
专题 | 沈阳自动化研究所_光电信息技术研究室 |
通讯作者 | Chen M(陈牧); Zhao HC(赵怀慈) |
作者单位 | 1.Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang, 110016, China 2.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, 110016, China 3.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, 110169, China 4.University of Chinese Academy of Sciences, Beijing, 100049, China |
推荐引用方式 GB/T 7714 | Chen M,Zhao HC,Liu PF. Monocular 3d object detection based on uncertainty prediction of keypoints[J]. Machines,2022,10(1):1-16. |
APA | Chen M,Zhao HC,&Liu PF.(2022).Monocular 3d object detection based on uncertainty prediction of keypoints.Machines,10(1),1-16. |
MLA | Chen M,et al."Monocular 3d object detection based on uncertainty prediction of keypoints".Machines 10.1(2022):1-16. |
入库方式: OAI收割
来源:沈阳自动化研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。