Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/donghwijung/lorcon-lo
LoRCoN-LO: Long-term Recurrent Convolutional Network-based LiDAR Odometry
https://github.com/donghwijung/lorcon-lo
deep-learning lcrn lidar-odometry paper rellis-3d robotics
Last synced: 2 days ago
JSON representation
LoRCoN-LO: Long-term Recurrent Convolutional Network-based LiDAR Odometry
- Host: GitHub
- URL: https://github.com/donghwijung/lorcon-lo
- Owner: donghwijung
- License: mit
- Created: 2022-03-10T11:28:08.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2024-02-14T07:48:05.000Z (12 months ago)
- Last Synced: 2024-02-14T08:39:03.887Z (12 months ago)
- Topics: deep-learning, lcrn, lidar-odometry, paper, rellis-3d, robotics
- Language: Python
- Homepage:
- Size: 33.2 KB
- Stars: 8
- Watchers: 2
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# LoRCoN-LO: Long-term Recurrent Convolutional Network-based LiDAR Odometry
Video [[link]](https://youtu.be/_Ld58Rn_y-s) Paper [[link]](https://arxiv.org/pdf/2303.11853.pdf)
## Download datasets
KITTI dataset (http://www.cvlibs.net/datasets/kitti/eval_odometry.php).
- velodyne laser data, calibration files, ground truth posesRellis-3D dataset (https://github.com/unmannedlab/RELLIS-3D).
- SemanticKITTI Format, Scan Poses files## Setup directories
```bash
bash setup_directories.sh
```
Enter dataset name
```
KITTI # KITTI dataset
Rellis-3D # Rellis-3D dataset
```## Move datasets
### KITTI
Move calib files (from 00 to 10)
```bash
mv data_odometry_calib/dataset/sequences/00/calib.txt data/KITTI/calib/00.txt
```
Move pose files (from 00 to 10)
```bash
mv data_odometry_poses/dataset/poses/00.txt data/KITTI/pose/00.txt
```
Move scan files (from 00 to 10)
```bash
mv data_odometry_velodyne/dataset/sequences/00/velodyne data/KITTI/scan/00/
```### Rellis-3D
Move pose files (from 00 to 04)
```bash
mv Rellis_3D_lidar_poses_20210614/Rellis_3D/000000/poses.txt data/Rellis-3D/pose/00.txt
```
Move scan files (from 00 to 04)
```bash
mv Rellis_3D_os1_cloud_node_kitti_bin/Rellis_3D/000000/os1_cloud_node_kitti_bin data/Rellis-3D/scan/00/
```## Setup the environment
```bash
conda create -n LoRCoN-LO python=3.8
conda activate LoRCoN-LO
pip install -r requirements.txt
```## Change the config file
Change the config file (`config/config.yaml`) depending on your directory configuration.## Pre-process
- transform ground truth poses from cam to vel
```bash
python preprocess/transform_poses_cam_to_vel.py
```## Compute relative poses
```bash
python preprocess/relative_pose_calculator.py
```## Generate input data
```bash
python preprocess/gen_data.py
```## Train and Test
```bash
python train.py
python test.py
```## Pre-trained models
KITTI model wass trained with 00 to 08 sequences.- [Download](https://drive.google.com/file/d/1pw8I5LrH5BI5_Oo8G7rhdiXX-vJlauIi/view?usp=sharing)
Rellis-3D model was trained with 00 to 03 sequences.
- [Download](https://drive.google.com/file/d/1edg5MsBRWxS7Q-slpvR3zbS8mIroLTUB/view?usp=sharing)
When downloading and running the model, please modify the checkpoint related code in `confg/config.yaml`.
## Paper
```
@inproceedings{jung2023lorcon,
title={LoRCoN-LO: Long-term Recurrent Convolutional Network-based LiDAR Odometry},
author={Jung, Donghwi and Cho, Jae-Kyung and Jung, Younghwa and Shin, Soohyun and Kim, Seong-Woo},
booktitle={2023 International Conference on Electronics, Information, and Communication (ICEIC)},
pages={1--4},
year={2023},
organization={IEEE}
}
```## Acknowledgement
The input data generation module is adapted from [Overlapnet](https://github.com/PRBonn/OverlapNet).## License
Copyright 2022, Donghwi Jung, Jae-Kyung Cho, Younghwa Jung, Soohyun Shin, Seong-Woo Kim, Autonomous Robot Intelligence Lab, Seoul National University.
This project is free software made available under the MIT License. For details see the LICENSE file.