4DenoiseNet: Adverse Weather Denoising From Adjacent Point Clouds

Loading...
Thumbnail Image
Access rights
openAccess
Journal Title
Journal ISSN
Volume Title
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Date
2023-01-01
Major/Subject
Mcode
Degree programme
Language
en
Pages
8
456-463
Series
IEEE Robotics and Automation Letters, Volume 8, issue 1
Abstract
Reliable point cloud data is essential for perception tasks e.g. in robotics and autonomous driving applications. Adverse weather causes a specific type of noise to light detection and ranging (LiDAR) sensor data, which degrades the quality of the point clouds significantly. To address this issue, this letter presents a novel point cloud adverse weather denoising deep learning algorithm (4DenoiseNet). Our algorithm takes advantage of the time dimension unlike deep learning adverse weather denoising methods in the literature. It performs about 10% better in terms of intersection over union metric compared to the previous work and is more computationally efficient. These results are achieved on our novel SnowyKITTI dataset, which has over 40000 adverse weather annotated point clouds. Moreover, strong qualitative results on the Canadian Adverse Driving Conditions dataset indicate good generalizability to domain shifts and to different sensor intrinsics.
Description
Funding Information: This work was supported in part by Henry Ford Foundation Finland and in part by the Helsinki Institute of Physics. Publisher Copyright: © 2022 IEEE.
Keywords
AI-based methods, computer vision for transportation, deep learning for visual perception, intelligent transportation systems, visual learning
Other note
Citation
Seppanen, A, Ojala, R & Tammi, K 2023, ' 4DenoiseNet: Adverse Weather Denoising From Adjacent Point Clouds ', IEEE Robotics and Automation Letters, vol. 8, no. 1, pp. 456-463 . https://doi.org/10.1109/LRA.2022.3227863