4DenoiseNet: Adverse Weather Denoising from Adjacent Point Clouds

TL;DR

This paper introduces 4DenoiseNet, a deep learning algorithm that leverages temporal information to effectively denoise LiDAR point clouds affected by adverse weather, improving accuracy and efficiency.

Contribution

It presents a novel 4D deep learning approach for adverse weather LiDAR denoising, utilizing temporal data and demonstrating superior performance on new datasets.

Findings

10% better intersection over union metric

More computationally efficient than previous methods

Good generalization to different datasets and sensor conditions

Abstract

Reliable point cloud data is essential for perception tasks \textit{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.