Home

Awesome

LiDARSeg3D

A repository for LiDAR 3D semantic segmentation in autonomous driving scenarios.

Also the official implementations of our ECCV 2022 paper (Self-Distillation for Robust LiDAR Semantic Segmentation in Autonomous Driving) and CVPR 2023 paper (MSeg3D: Multi-modal 3D Semantic Segmentation for Autonomous Driving).

News

<!-- - [2022-07-14] Initial release for the implementation of SDSeg3D. --> <p align="center"> <img src='docs/semnusc_leaderboard.png' align="center" height="240px"> </p> <!-- ## Contact Any questions or suggestions are welcome! Jiale Li [jialeli@zju.edu.cn](mailto:jialeli@zju.edu.cn) (ZJU), and Hang Dai [hang.dai.cs@gmail.com](mailto:hang.dai.cs@gmail.com) (MBZUAI) -->

Highlights

Methods

MSeg3D

MSeg3D: Multi-modal 3D Semantic Segmentation for Autonomous Driving
Jiale Li, Hang Dai, Hao Han, and Yong Ding

SDSeg3D

Self-Distillation for Robust LiDAR Semantic Segmentation in Autonomous Driving
Jiale Li, Hang Dai, and Yong Ding

Use LiDARSeg3D

Installation

Please follow INSTALL to set up libraries needed for distributed training and sparse convolution.

Benchmark Evaluation and Training

Please refer to GETTING_START to prepare the data in advance. Then follow the instruction there to play with the segmentation configurations included in configs.

Acknowlegement

This project is mainly constructed on CenterPoint as well as multiple great opensourced codebases. We list some notable examples below.

Citation

@inproceedings{mseg3d_cvpr2023,
author    = {Jiale Li and
            Hang Dai and
            Hao Han and
            Yong Ding},
title     = {MSeg3D: Multi-modal 3D Semantic Segmentation for Autonomous Driving},
booktitle = {CVPR},
pages     = {21694--21704},
year      = {2023},
}

@inproceedings{sdseg3d_eccv2022,
author    = {Jiale Li and
            Hang Dai and
            Yong Ding},
title     = {Self-Distillation for Robust {LiDAR} Semantic Segmentation in Autonomous Driving},
booktitle = {ECCV},
pages     = {659--676},
year      = {2022},
}