ICLR 2024poster10 citations

ReSimAD: Zero-Shot 3D Domain Transfer for Autonomous Driving with Source Reconstruction and Target Simulation

Bo Zhang, Xinyu Cai, Jiakang Yuan, Donglin Yang, Jianfei Guo, Xiangchao Yan, Renqiu Xia, Botian Shi

Abstract

Domain shifts such as sensor type changes and geographical situation variations are prevalent in Autonomous Driving (AD), which poses a challenge since AD model relying on the previous domain knowledge can be hardly directly deployed to a new domain without additional costs. In this paper, we provide a new perspective and approach of alleviating the domain shifts, by proposing a Reconstruction-Simulation-Perception (ReSimAD) scheme. Specifically, the implicit reconstruction process is based on the knowledge from the previous old domain, aiming to convert the domain-related knowledge into domain-invariant representations, e.g., 3D scene-level meshes. Besides, the point clouds simulation process of multiple new domains is conditioned on the above reconstructed 3D meshes, where the target-domain-like simulation samples can be obtained, thus reducing the cost of collecting and annotating new-domain data for the subsequent perception process. For experiments, we consider different cross-domain situations such as Waymo-to-KITTI, Waymo-to-nuScenes, etc, to verify the zero-shot target-domain perception using ReSimAD. Results demonstrate that our method is beneficial to boost the domain generalization ability, even promising for 3D pre-training. Code and simulated points are available at: https://github.com/PJLab-ADG/3DTrans

Autonomous Driving3D Domain TransferZero-shot 3D Detection
BibTeX
@inproceedings{
zhang2024resimad,
title={ReSim{AD}: Zero-Shot 3D Domain Transfer for Autonomous Driving with Source Reconstruction and Target Simulation},
author={Bo Zhang and Xinyu Cai and Jiakang Yuan and Donglin Yang and Jianfei Guo and Xiangchao Yan and Renqiu Xia and Botian Shi and Min Dou and Tao Chen and Si Liu and Junchi Yan and Yu Qiao},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=1d2cLKeNgY}
}
ReSimAD: Zero-Shot 3D Domain Transfer for Autonomous Driving with Source Reconstruction and Target Simulation · ICLR 2024