Scene L004.ply from Toronto3D dataset is used.
- Load .ply file and parse properties
- Filter points with label 0 from ply files
- Filter points by height and beam angle
- Filter points by min z difference(A Practical Point Cloud Based Road Curb Detection Method for Autonomous Vehicle)
- Save result in .las file
| Imported cloud | Curb segmentation on crossroad |
|---|---|
 |
 |
- Python 3.7
- Open3d
- Pyyaml
- Numpy
- pylas
pip install -r requirements.txt
usage: Road curb cloud segmentation [-h] [--output_dir OUTPUT_DIR] input_file
positional arguments:
input_file path to .ply file
optional arguments:
-h, --help show this help message and exit
--output_dir OUTPUT_DIR
path to output .las file