Simulation-to-Reality Domain Adaptation for Offline 3D Object Annotation on Pointclouds with Correlation Alignment

Weishuang Zhang, B. Kiran, Thomas Gauthier, Yanis Mazouz, Theo Steger

2022

Abstract

Annotating objects with 3D bounding boxes in LiDAR pointclouds is a costly human driven process in an autonomous driving perception system. In this paper, we present a method to semi-automatically annotate real-world pointclouds collected by deployment vehicles using simulated data. We train a 3D object detector model on labeled simulated data from CARLA jointly with real world pointclouds from our target vehicle. The supervised object detection loss is augmented with a CORAL loss term to reduce the distance between labeled simulated and unlabeled real pointcloud feature representations. The goal here is to learn representations that are invariant to simulated (labeled) and real-world (unlabeled) target domains. We also provide an updated survey on domain adaptation methods for pointclouds.

Paper Citation

in Harvard Style

Zhang W., Kiran B., Gauthier T., Mazouz Y. and Steger T. (2022). Simulation-to-Reality Domain Adaptation for Offline 3D Object Annotation on Pointclouds with Correlation Alignment. In Proceedings of the 2nd International Conference on Image Processing and Vision Engineering - Volume 1: IMPROVE, ISBN 978-989-758-563-0, pages 142-149. DOI: 10.5220/0011059200003209

in Bibtex Style

@conference{improve22,
author={Weishuang Zhang and B. Kiran and Thomas Gauthier and Yanis Mazouz and Theo Steger},
title={Simulation-to-Reality Domain Adaptation for Offline 3D Object Annotation on Pointclouds with Correlation Alignment},
booktitle={Proceedings of the 2nd International Conference on Image Processing and Vision Engineering - Volume 1: IMPROVE,},
year={2022},
pages={142-149},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011059200003209},
isbn={978-989-758-563-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Image Processing and Vision Engineering - Volume 1: IMPROVE,
TI - Simulation-to-Reality Domain Adaptation for Offline 3D Object Annotation on Pointclouds with Correlation Alignment
SN - 978-989-758-563-0
AU - Zhang W.
AU - Kiran B.
AU - Gauthier T.
AU - Mazouz Y.
AU - Steger T.
PY - 2022
SP - 142
EP - 149
DO - 10.5220/0011059200003209