Robust 3D Object Detection in Cold Weather Conditions

TL;DR

This paper addresses the challenge of gas exhaust condensation in cold weather affecting LiDAR-based object detection by proposing data augmentation, a novel loss, and a generation method to improve robustness without changing network architecture.

Contribution

It introduces a gas exhaust data generation technique, a point cloud augmentation process, and a new training loss to enhance detection robustness under adverse weather conditions.

Findings

Significant improvement in detection robustness against gas exhaust noise.

Method works with both grid-based and point-based detectors.

No change in inference time or network architecture.

Abstract

Adverse weather conditions can negatively affect LiDAR-based object detectors. In this work, we focus on the phenomenon of vehicle gas exhaust condensation in cold weather conditions. This everyday effect can influence the estimation of object sizes, orientations and introduce ghost object detections, compromising the reliability of the state of the art object detectors. We propose to solve this problem by using data augmentation and a novel training loss term. To effectively train deep neural networks, a large set of labeled data is needed. In case of adverse weather conditions, this process can be extremely laborious and expensive. We address this issue in two steps: First, we present a gas exhaust data generation method based on 3D surface reconstruction and sampling which allows us to generate large sets of gas exhaust clouds from a small pool of labeled data. Second, we introduce a point cloud augmentation process that can be used to add gas exhaust to datasets recorded in good weather conditions. Finally, we formulate a new training loss term that leverages the augmented point cloud to increase object detection robustness by penalizing predictions that include noise. In contrast to other works, our method can be used with both grid-based and point-based detectors. Moreover, since our approach does not require any network architecture changes, inference times remain unchanged. Experimental results on real data show that our proposed method greatly increases robustness to gas exhaust and noisy data.