FCGF
Fully Convolutional Geometric Features: Fast and accurate 3D features for registration and correspondence.
File Explorer
Download Latest Version (.zip)- 1.png
- 2.png
- 3.png
- 3_1.png
- 3_2.png
- 4.png
- demo.png
- fig4_dist_thresh.txt
- fig4_inlier_thresh.txt
- fps_acc.png
- kitchen_0.png
- kitchen_1.png
- table.png
- text_scene000.gif
- test_3dmatch.txt
- test_kitti.txt
- train_3dmatch.txt
- train_kitti.txt
- val_3dmatch.txt
- val_kitti.txt
- __init__.py
- data_loaders.py
- eval.py
- metrics.py
- timer.py
- trainer.py
- transforms.py
- __init__.py
- common.py
- residual_block.py
- resunet.py
- simpleunet.py
- benchmark_3dmatch.py
- benchmark_util.py
- download_3dmatch_test.sh
- download_datasets.sh
- test_kitti.py
- train_fcgf_kitti.sh
- __init__.py
- file.py
- misc.py
- pointcloud.py
- trajectory.py
- transform_estimation.py
- visualization.py
- check_pretrained.py
- convert_me_to_wcn.py
- dataset.py
- eval_3dmatch.py
- features.py
- loss.py
- perm_sweep.py
- README.md
- smoke.py
- train.py
- verify_conv.py
- .gitignore
- .style.yapf
- benchmark.py
- config.py
- demo.py
- LICENSE
- README.md
- requirements.txt
- train.py
# Installation Guide
git clone https://github.com/chrischoy/FCGF
Downloads the entire project code from GitHub to your computer.
cd FCGF
Moves into the project folder you just downloaded.
2. Official Install Script
Easy Recommended- Python 3 Python is required to use pip.
pip install torch==2.5.1 --index-url https://download.pytorch.org/whl/cu124
Installs the package published on PyPI directly โ no need to clone the source.
pip install https://github.com/NVlabs/WarpConvNet/releases/download/v1.7.11/warpconvnet-1.7.11+torch2.5cu124-cp312-cp312-linux_x86_64.whl
Installs the package published on PyPI directly โ no need to clone the source.
pip install scipy plyfile # + open3d for the demo viz
Installs the package published on PyPI directly โ no need to clone the source.
pip install torch-scatter -f https://data.pyg.org/whl/torch-2.5.1+cu124.html
Installs the package published on PyPI directly โ no need to clone the source.
Pulled directly from this repo's README.
3. Python
Easypip install torch==2.5.1 --index-url https://download.pytorch.org/whl/cu124
Installs the package published on PyPI directly โ no need to clone the source.
pip install https://github.com/NVlabs/WarpConvNet/releases/download/v1.7.11/warpconvnet-1.7.11+torch2.5cu124-cp312-cp312-linux_x86_64.whl
Installs the package published on PyPI directly โ no need to clone the source.
pip install scipy plyfile # + open3d for the demo viz
Installs the package published on PyPI directly โ no need to clone the source.
pip install torch-scatter -f https://data.pyg.org/whl/torch-2.5.1+cu124.html
Installs the package published on PyPI directly โ no need to clone the source.
Pulled directly from this repo's README.