Towards Minimizing the LiDAR Sim-to-Real Domain Shift: Object-Level Local Domain Adaptation for 3D Point Clouds of Autonomous Vehicles

Sebastian Huch, Markus Lienkamp

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

Perception algorithms for autonomous vehicles demand large, labeled datasets. Real-world data acquisition and annotation costs are high, making synthetic data from simulation a cost-effective option. However, training on one source domain and testing on a target domain can cause a domain shift attributed to local structure differences, resulting in a decrease in the model’s performance. We propose a novel domain adaptation approach to address this challenge and to minimize the domain shift between simulated and real-world LiDAR data. Our approach adapts 3D point clouds on the object level by learning the local characteristics of the target domain. A key feature involves downsampling to ensure domain invariance of the input data. The network comprises a state-of-the-art point completion network combined with a discriminator to guide training in an adversarial manner. We quantify the reduction in domain shift by training object detectors with the source, target, and adapted datasets. Our method successfully reduces the sim-to-real domain shift in a distribution-aligned dataset by almost 50%, from 8.63% to 4.36% 3D average precision. It is trained exclusively using target data, making it scalable and applicable to adapt point clouds from any source domain.

Original languageEnglish
Article number9913
JournalSensors (Switzerland)
Volume23
Issue number24
DOIs
StatePublished - Dec 2023

Keywords

  • LIDAR
  • autonomous vehicles
  • deep learning
  • domain adaptation
  • object detection
  • point cloud
  • synthetic data

Fingerprint

Dive into the research topics of 'Towards Minimizing the LiDAR Sim-to-Real Domain Shift: Object-Level Local Domain Adaptation for 3D Point Clouds of Autonomous Vehicles'. Together they form a unique fingerprint.

Cite this