---
license: mit
tags:
- point-cloud
- 3d-registration
- geometric-features
- pytorch
- minkowski-engine
---

# Fully Convolutional Geometric Features (FCGF)

Pretrained weights for **Fully Convolutional Geometric Features**, ICCV 2019.

- Paper: [Fully Convolutional Geometric Features](https://node1.chrischoy.org/data/publications/fcgf/fcgf.pdf)
- Code: [chrischoy/FCGF](https://github.com/chrischoy/FCGF)

FCGF extracts dense 3D geometric features in a single forward pass through a 3D fully-convolutional network, built on [MinkowskiEngine](https://github.com/NVIDIA/MinkowskiEngine). These checkpoints are the official weights released with the paper, previously hosted on `node1.chrischoy.org` and mirrored here.

## Model Zoo

All models use the `ResUNetBN2C` backbone.

| File                                          | Normalized Feature | Dataset | Voxel Size    | Feature Dimension | Performance                |
|:-----------------------------------------------|:-------------------:|:-------:|:-------------:|:-----------------:|:--------------------------:|
| [2019-08-19_06-17-41.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/2019-08-19_06-17-41.pth)                             | True  | 3DMatch | 2.5cm (0.025) | 32 | FMR: 0.9578 +- 0.0272      |
| [2019-09-18_14-15-59.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/2019-09-18_14-15-59.pth)                             | True  | 3DMatch | 2.5cm (0.025) | 16 | FMR: 0.9442 +- 0.0345      |
| [2019-08-16_19-21-47.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/2019-08-16_19-21-47.pth)                             | True  | 3DMatch | 5cm   (0.05)  | 32 | FMR: 0.9372 +- 0.0332      |
| [2019-07-31_19-30-19.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/2019-07-31_19-30-19.pth)                             | False | KITTI   | 20cm  (0.2)   | 32 | RTE: 0.0534m, RRE: 0.1704° |
| [2019-07-31_19-37-00.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/2019-07-31_19-37-00.pth)                             | False | KITTI   | 30cm  (0.3)   | 32 | RTE: 0.0607m, RRE: 0.2280° |
| [KITTI-v0.3-ResUNetBN2C-conv1-5-nout16.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/KITTI-v0.3-ResUNetBN2C-conv1-5-nout16.pth) | True  | KITTI   | 30cm  (0.3)   | 16 | RTE: 0.0670m, RRE: 0.2295° |
| [KITTI-v0.3-ResUNetBN2C-conv1-5-nout32.pth](https://huggingface.co/chrischoy/FCGF/resolve/main/KITTI-v0.3-ResUNetBN2C-conv1-5-nout32.pth) | True  | KITTI   | 30cm  (0.3)   | 32 | RTE: 0.0639m, RRE: 0.2253° |

`redkitchen-20.ply` is the sample point cloud used by [`demo.py`](https://github.com/chrischoy/FCGF/blob/master/demo.py) in the FCGF repo.

## Usage

```python
import torch
from urllib.request import urlretrieve

urlretrieve(
    "https://huggingface.co/chrischoy/FCGF/resolve/main/2019-09-18_14-15-59.pth",
    "ResUNetBN2C-16feat-3conv.pth")

checkpoint = torch.load("ResUNetBN2C-16feat-3conv.pth")
```

See [chrischoy/FCGF](https://github.com/chrischoy/FCGF) for the full model definition (`ResUNetBN2C`) and feature extraction code.

## Citation

```bibtex
@inproceedings{FCGF2019,
    author = {Christopher Choy and Jaesik Park and Vladlen Koltun},
    title = {Fully Convolutional Geometric Features},
    booktitle = {ICCV},
    year = {2019},
}
```