Improving LiDAR Semantic Segmentation on Minority Classes and Generalization Capability for Autonomous Driving

Chiao Hua Tseng, Yu Ting Lin, Wen Chieh Lin, Chieh Chih Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

LiDARs have emerged as an important sensor in autonomous driving systems because they offer more accurate geometric measurements than cameras and radars. Therefore, LiDARs have been commonly combined with cameras or radars to tackle many perception problems in autonomous driving, such as object detection, semantic segmentation, or navigation. For semantic segmentation of LiDAR data, due to the class imbalance issue of large-scale scene, there is a performance gap between majority classes and minority classes of large-scale dataset. The minority classes usually include the crucial classes to the autonomous driving, such as 'person', 'motorcyclist', 'traffic-sign'. To improve the performance of minority classes, we adopt U-Net++ as the architecture, KPConv as convolution operator, and use both dice loss and cross entropy as loss functions. We get 5.1% mIoU improvement on SemanticKITTI of all classes and 9.5% mIoU improvement of minority classes. Moreover, due to the different resolution of LiDAR sensors, we show the generalization capability of our model by training it on 64-beam dataset and testing on 32-beam and 128-beam dataset. We get 3.3% mIoU improvement on 128-beam dataset and 1.9% mIoU improvement on 32-beam dataset.

Original languageEnglish
Title of host publicationProceedings - 2022 International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages131-136
Number of pages6
ISBN (Electronic)9798350399509
DOIs
StatePublished - 2022
Event27th International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022 - Tainan, Taiwan
Duration: 1 Dec 20223 Dec 2022

Publication series

NameProceedings - 2022 International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022

Conference

Conference27th International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022
Country/TerritoryTaiwan
CityTainan
Period1/12/223/12/22

Keywords

  • autonomous driving
  • deep learning
  • generalization capability
  • LiDAR semantic segmentation
  • minority class

Fingerprint

Dive into the research topics of 'Improving LiDAR Semantic Segmentation on Minority Classes and Generalization Capability for Autonomous Driving'. Together they form a unique fingerprint.

Cite this