https://github.com/qcraftai/pillar-motion
Self-Supervised Pillar Motion Learning for Autonomous Driving (CVPR 2021)
https://github.com/qcraftai/pillar-motion
autonomous-driving lidar-point-cloud nuscenes point-cloud-motion self-supervised-learning
Last synced: 7 months ago
JSON representation
Self-Supervised Pillar Motion Learning for Autonomous Driving (CVPR 2021)
- Host: GitHub
- URL: https://github.com/qcraftai/pillar-motion
- Owner: qcraftai
- License: other
- Created: 2021-03-28T03:11:06.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2023-07-30T02:45:31.000Z (about 2 years ago)
- Last Synced: 2024-08-01T03:39:10.114Z (about 1 year ago)
- Topics: autonomous-driving, lidar-point-cloud, nuscenes, point-cloud-motion, self-supervised-learning
- Language: Python
- Homepage:
- Size: 23.1 MB
- Stars: 119
- Watchers: 10
- Forks: 12
- Open Issues: 12
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
## Self-Supervised Pillar Motion Learning for Autonomous Driving
[Chenxu Luo](https://chenxuluo.github.io/), [Xiaodong Yang](https://xiaodongyang.org/), [Alan Yuille](https://www.cs.jhu.edu/~ayuille/)
Self-Supervised Pillar Motion Learning for Autonomous Driving, CVPR 2021
[[Paper]](https://arxiv.org/pdf/2104.08683.pdf) [[Poster]](poster.pdf) [[YouTube]](https://youtu.be/Y00ujpmauUU)
## Getting Started
### Installation
Install [PyTorch](https://pytorch.org/), [PyTorch3D](https://github.com/facebookresearch/pytorch3d), [Apex](https://github.com/NVIDIA/apex), [nuScenes Devkit](https://github.com/nutonomy/nuscenes-devkit)
### Data Preparation
```
python tools/create_data nuscenes_data_prep --root_path /path/to/nuscenes
```
Our optical flow model used for the cross-sensor regularization is available [here](https://drive.google.com/file/d/1H-UdwMrmIDJNs_tGcGXL2c_UqZFuUaIZ/view?usp=sharing).
### Training
```
python -m torch.distributed.launch --nproc_per_node=8 ./tools/train.py configs/nusc_pillarmotion.py --work_dir experiments/pillarmotion/
```
## Citation
Please cite the following paper if this repo helps your research:
```bibtex
@InProceedings{Luo_2021_CVPR,
author = {Luo, Chenxu and Yang, Xiaodong and Yuille, Alan},
title = {Self-Supervised Pillar Motion Learning for Autonomous Driving},
booktitle = {Computer Vision and Pattern Recognition (CVPR)},
year = {2021}
}
```
## License
Copyright (C) 2021 QCraft. All rights reserved. Licensed under the [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode) (Attribution-NonCommercial-ShareAlike 4.0 International). The code is released for academic research use only. For commercial use, please contact [business@qcraft.ai](business@qcraft.ai).