zhouzheyuan / sgm3d

[RA-L 2022] SGM3D: Stereo Guided Monocular 3D Object Detection

SGM3D

SGM3D: Stereo Guided Monocular 3D Object Detection,
Zheyuan Zhou, Liang Du, Xiaoqing Ye, Zhikang Zou, Xiao Tan, Li Zhang, Xiangyang Xue, Jianfeng Feng
RA-L 2022

Overview

• Changelog
• Model Zoo
• Installation
• Getting Started
• Citation

Changelog

[2022-07-14] Paper is received by RA-L.

[2022-06-24] Code is released.

[2021-12-03] Paper is released on Arxiv.

Model Zoo

KITTI 3D Object Detection Baselines

Selected supported methods are shown in the below table. The results are the 3D detection performance of car class on the val set of KITTI dataset.

	Easy@R40	Mod.@R40	Hard@R40	download
SGM3D	25.95	17.44	15.36	model-292M

Installation

Please refer to INSTALL.md for the installation of OpenPCDet.

Getting Started

Please refer to GETTING_STARTED.md to learn more usage about this project.

Acknowledgements

SGM3D is based on OpenPCDet.

Citation

If you find this project useful in your research, please consider cite:

@article{zhou2022sgm3d,
  title={SGM3D: Stereo Guided Monocular 3D Object Detection}, 
  author={Zhou, Zheyuan and Du, Liang and Ye, Xiaoqing and Zou, Zhikang and Tan, Xiao and Zhang, Li and Xue, Xiangyang and Feng, Jianfeng},
  journal={IEEE Robotics and Automation Letters}, 
  year={2022},
  volume={7},
  number={4},
  pages={10478-10485},
  doi={10.1109/LRA.2022.3191849}}