EfficientLPS is a state-of-the-art top-down approach for LiDAR panoptic segmentation, where the goal is to assign semantic labels (e.g., car, road, tree and so on) to every point in the input LiDAR point cloud as well as instance labels (e.g. an id of 1, 2, 3, etc) to points belonging to thing classes.
This repository contains the PyTorch implementation of our T-RO'2021 paper EfficientLPS: Efficient LiDAR Panoptic Segmentation. The repository builds on mmdetection and gen-efficientnet-pytorch codebases.
If you find the code useful for your research, please consider citing our paper:
@article{sirohi2021efficientlps,
title={Efficientlps: Efficient lidar panoptic segmentation},
author={Sirohi, Kshitij and Mohan, Rohit and B{\"u}scher, Daniel and Burgard, Wolfram and Valada, Abhinav},
journal={IEEE Transactions on Robotics},
volume={38},
number={3},
pages={1894--1914},
year={2021},
publisher={IEEE}
}
EfficientPS is also featured in the OpenDR toolkit.
https://rl.uni-freiburg.de/research/lidar-panoptic
- Linux
- Python 3.7
- PyTorch 1.7
- CUDA 10.2
- GCC 7 or 8
IMPORTANT NOTE: These requirements are not necessarily mandatory. However, we have only tested the code under the above settings and cannot provide support for other setups.
a. Create a conda virtual environment from the provided environment.yml and activate it.
git clone https://github.com/robot-learning-freiburg/EfficientLPS.git
cd EfficientLPS
conda env create -n efficientLPS_env --file=environment.yml
conda activate efficientLPS_envb. Install all other dependencies using pip:
pip install -r requirements.txtc. Install EfficientNet implementation
cd efficientNet
python setup.py developd. Install EfficientLPS implementation
cd ..
python setup.py develop- Download the dataset from here.
- Expected Folder Structure:
/kitti/dataset/
└── sequences/
├── 00/
│ ├── poses.txt
│ ├── image_2/
│ ├── image_3/
│ ├── labels/
│ │ ├ 000000.label
│ │ └ 000001.label
│ └── velodyne/
│ ├ 000000.bin
│ └ 000001.bin
├── 01/
├── 02/
.
.
.
└── 21/
- In config file set the data_root variable to your own full path of /kitti/dataset/
Edit the config file appropriately in configs folder and run the following command:
./tools/dist_train.sh efficientPS_multigpu_sample.py ${GPU_NUM} --work_dir ../work_dirs/checkpoints --validate
- --resume_from ${CHECKPOINT_FILE}: Resume from a previous checkpoint file.
Run the following command for evaluation:
./tools/dist_test.sh efficientPS_multigpu_sample.py ${CHECKPOINT_FILE} ${GPU_NUM} --eval panoptic
| Dataset | Model |
|---|---|
| SemanticKITTI | Download |
We have used utility functions from other open-source projects. We especially thank the authors of:
For academic usage, the code is released under the GPLv3 license. For any commercial purpose, please contact the authors.