Voxblox Python Bindings

This small library provides python Bindings for the Voxblox library¹. The original code has not been modified at all(even more, it's just a hard dependency of this project) and it's only exposing some itegration methods(for TSDF reconstruction) in Python, this make experimentation much easier than relying on the full ROS ecosystem and bagfiles.

Example results

In case you wonder why that weird wall is on the output mesh, that's an already known bug of the original voxblox library... I'm sorry, it's not because of the python bindings.

Installation

Dependencies

The only dependencies for this library it's a propper C++ compiler, if you are using ubuntu-based distribution this command will give you all the necessary dependencies:

C++ dependencies

sudo apt-get update && apt-get install build-essential cmake libprotobuf-dev protobuf-compiler

Python dependencies

sudo apt-get update && apt-get install python3 python3-dev python3-pip

Installing the python bindings

Just clone this repo:

git clone --recurse-submodules https://https://github.com/PRBonn/voxblox_pybind

And install the python bindings:

cd voxblox_pybind
make install

Usage

The only classes that are exposed through the Python API are:

Considering that you have already a dataset that provide you with pointclouds, the usage of the python bindings are quite straight forward:

# It's your responsability to provide this dataset implementation
dataset = SimpleDataset()

# Pick some parameters
voxel_size = 0.1
sdf_trunc = 3 * voxel_size

# Run fusion pipeline
tsdf_volume = SimpleTsdfIntegrator(voxel_size, sdf_trunc)
for idx in range(len(dataset)):
    scan, pose = dataset[idx]
    tsdf_volume.integrate(scan, pose)

# Get the output mesh
vertices, triangles = tsdf_volume.extract_triangle_mesh()
mesh = o3d.geometry.TriangleMesh(
    o3d.utility.Vector3dVector(vertices),
    o3d.utility.Vector3iVector(triangles),
)
mesh.compute_vertex_normals()
o3d.visualization.draw_geometries([mesh])

A full self-conctained example can be found at test_voxblox.py and a more complete pipeline at kitti_pipeline.py

Credits

This code has been created to support the development of VDBFusion. If you use our code in your academic work, please cite the corresponding paper and the original voxblox paper:

@article{vizzo2022sensors,
  author         = {Vizzo, Ignacio and Guadagnino, Tiziano and Behley, Jens and Stachniss, Cyrill},
  title          = {VDBFusion: Flexible and Efficient TSDF Integration of Range Sensor Data},
  journal        = {Sensors},
  volume         = {22},
  year           = {2022},
  number         = {3},
  article-number = {1296},
  url            = {https://www.mdpi.com/1424-8220/22/3/1296},
  issn           = {1424-8220},
  doi            = {10.3390/s22031296}
}

Original voxblox paper:

@inproceedings{oleynikova2017voxblox,
  author={Oleynikova, Helen and Taylor, Zachary and Fehr, Marius and Siegwart, Roland and  Nieto, Juan},
  booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  title={Voxblox: Incremental 3D Euclidean Signed Distance Fields for On-Board MAV Planning},
  year={2017}
}

All voxblox documentation can be found on the original readthedocs page ↩

PRBonn/voxblox_pybind