https://github.com/prbonn/auto-mos
Automatic Labeling to Generate Training Data for Online LiDAR-based Moving Object Segmentation
https://github.com/prbonn/auto-mos
Last synced: 6 months ago
JSON representation
Automatic Labeling to Generate Training Data for Online LiDAR-based Moving Object Segmentation
- Host: GitHub
- URL: https://github.com/prbonn/auto-mos
- Owner: PRBonn
- License: mit
- Created: 2022-03-02T08:54:18.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2022-08-13T19:06:46.000Z (over 3 years ago)
- Last Synced: 2023-11-17T04:03:13.857Z (about 2 years ago)
- Language: Python
- Size: 1.2 MB
- Stars: 79
- Watchers: 10
- Forks: 5
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Auto-MOS: Automatic Labeling to Generate Training Data for Online LiDAR-based Moving Object Segmentation
This repo contains the code for our Auto-MOS, which automatically generates training data for LiDAR-based moving objects segmentation [PDF](http://arxiv.org/pdf/2201.04501).
### Table of Contents
1. [Introduction](#Auto-MOS:-Automatic-Labeling-to-Generate-Training-Data-for-Online-LiDAR-based-Moving-Object-Segmentation)
2. [Publication](#Publication)
3. [Logs](#Logs)
4. [Dependencies](#Dependencies)
5. [How to use](#How-to-use)
6. [Application](#Application)
7. [License](#License)
## Publication
If you use our implementation in your academic work, please cite the corresponding paper ([PDF](http://arxiv.org/pdf/2201.04501)):
@article{chen2022ral,
author = {X. Chen and B. Mersch and L. Nunes and R. Marcuzzi and I. Vizzo and J. Behley and C. Stachniss},
title = {{Automatic Labeling to Generate Training Data for Online LiDAR-Based Moving Object Segmentation}},
journal = {IEEE Robotics and Automation Letters (RA-L)},
year = 2022,
volume = 7,
number = 3,
pages = {6107-6114},
url = {http://arxiv.org/pdf/2201.04501},
issn = {2377-3766},
doi = {10.1109/LRA.2022.3166544}
}
## Logs
### Version 1.0
Note that, due to copyright and protection of our benchmark, this repo currently only provides the tracking and label generating parts of the proposed method.
For Odometry/LiDAR-SLAM we refer to our SuMa ([link](https://github.com/jbehley/SuMa)), refer dynamic removal to ERASOR ([link](https://github.com/LimHyungTae/ERASOR)), refer instance clustering to HDBSCAN ([link](https://github.com/scikit-learn-contrib/hdbscan)), and refer the LiDAR-MOS network to our LMNet ([link](https://github.com/PRBonn/LiDAR-MOS)).
## Dependencies
Before using our code, you need to install some libraries.
- System dependencies:
```bash
sudo apt-get update
sudo apt-get install -y python3-pip wget unzip
sudo -H pip3 install --upgrade pip
```
- Python dependencies (may also work with different versions than mentioned in the requirements file)
```bash
sudo -H pip3 install -r requirements.txt
```
## How to run
### Download data and intermediate results
To run the quick demo, please first download the data ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/kitti.zip)) extracting it to the `data` folder, and the intermediate instance results ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/instances.zip)) extracting it to the `results` folder.
To visualize the final results, you could also directly download the mos results ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/mos_predictions.zip)) and extract it into the `results` folder.
You could also download the data and intermediate results using command lines as follows:
- Download kitti demo dataset:
```bash
wget -P data/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/kitti.zip
unzip data/kitti.zip -d data
rm data/kitti.zip
```
- Download instance predictions:
```bash
wget -P results/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/instances.zip
unzip results/instances.zip -d results
rm results/instances.zip
```
- Download final mos predictions:
```bash
wget -P results/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/mos_predictions.zip
unzip results/mos_predictions.zip -d results
rm results/mos_predictions.zip
```
### Quick run
- To automatic generate the mos labels, one could directly run:
```bash
python3 auto-mos-tracking.py
```
- To visualize the mos results, one could directly run:
```bash
python3 vis_mos_results.py
```
To control the visualizer:
- press `n`: play next scan,
- press `b`: play previous scan,
- press `esc` or `q`: exits.
- To visualize the intermediate instance predictions, one could directly run:
```bash
python3 vis_instances.py
```
To control the visualizer:
- press `esc` or `q`: exits.
## License
This project is free software made available under the MIT License. For details see the LICENSE file.