Keypoint3D: Keypoint-Based and Anchor-Free 3D Object Detection for Autonomous Driving with Monocular Vision
-
Published:2023-02-22
Issue:5
Volume:15
Page:1210
-
ISSN:2072-4292
-
Container-title:Remote Sensing
-
language:en
-
Short-container-title:Remote Sensing
Author:
Li Zhen1ORCID, Gao Yuliang1, Hong Qingqing2, Du Yuren2, Serikawa Seiichi1, Zhang Lifeng1ORCID
Affiliation:
1. Graduate School of Engineering, Kyushu Institute of Technology, Kitakyushu 804-0015, Japan 2. College of Artificial Intelligence, Yangzhou University, Yangzhou 225012, China
Abstract
Autonomous driving has received enormous attention from the academic and industrial communities. However, achieving full driving autonomy is not a trivial task, because of the complex and dynamic driving environment. Perception ability is a tough challenge for autonomous driving, while 3D object detection serves as a breakthrough for providing precise and dependable 3D geometric information. Inspired by practical driving experiences of human experts, a pure visual scheme takes sufficient responsibility for safe and stable autonomous driving. In this paper, we proposed an anchor-free and keypoint-based 3D object detector with monocular vision, named Keypoint3D. We creatively leveraged 2D projected points from 3D objects’ geometric centers as keypoints for object modeling. Additionally, for precise keypoints positioning, we utilized a novel self-adapting ellipse Gaussian filter (saEGF) on heatmaps, considering different objects’ shapes. We tried different variations of DLA-34 backbone and proposed a semi-aggregation DLA-34 (SADLA-34) network, which pruned the redundant aggregation branch but achieved better performance. Keypoint3D regressed the yaw angle in a Euclidean space, which resulted in a closed mathematical space avoiding singularities. Numerous experiments on the KITTI dataset for a moderate level have proven that Keypoint3D achieved the best speed-accuracy trade-off with an average precision of 39.1% at 18.9 FPS on 3D cars detection.
Funder
Kyushu Institute of Technology SPRING Scholarship Awardee University Fellowship Founding Project for Innovation Creation in Science and Technology Fellowship Program
Subject
General Earth and Planetary Sciences
Reference58 articles.
1. A survey on 3D object detection methods for autonomous driving applications;Arnold;IEEE Trans. Intell. Transp. Syst.,2019 2. Connected vehicles: Solutions and challenges;Lu;IEEE Internet Things J.,2014 3. Abdollahi, A., Pradhan, B., Shukla, N., Chakraborty, S., and Alamri, A. (2020). Deep learning approaches applied to remote sensing datasets for road extraction: A state-of-the-art review. Remote Sens., 12. 4. Russell, B.J., Soffer, R.J., Ientilucci, E.J., Kuester, M.A., Conran, D.N., Arroyo-Mora, J.P., Ochoa, T., Durell, C., and Holt, J. (2023). The ground to space calibration experiment (G-SCALE): Simultaneous validation of UAV, airborne, and satellite imagers for Earth observation using specular targets. Remote Sens., 15. 5. Gagliardi, V., Tosti, F., Bianchini Ciampoli, L., Battagliere, M.L., D’Amato, L., Alani, A.M., and Benedetto, A. (2023). Satellite remote sensing and non-destructive testing methods for transport infrastructure monitoring: Advances, challenges and perspectives. Remote Sens., 15.
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|