ROS-VDBFusion: Flexible and Efficient TSDF Integration

A ROS C++ wrapper to the vdbfusion library for flexible and efficient TSDF Integration

Installation

OpenVDB

```sh

Install OpenVDB dependencies

sudo apt-get update && apt-get install --no-install-recommends -y \ libblosc-dev \ libboost-iostreams-dev \ libboost-system-dev \ libboost-system-dev \ libeigen3-dev

Install OpenVDB from source

git clone --depth 1 https://github.com/nachovizzo/openvdb.git -b nacho/vdbfusion && cd openvdb mkdir build && cd build && cmake -DCMAKEPOSITIONINDEPENDENTCODE=ON -DUSEZLIB=OFF .. && sudo make -j$(nproc) all install ```

VDBfusion

sh git clone --depth 1 https://github.com/PRBonn/vdbfusion.git && cd vdbfusion mkdir build && cd build && cmake .. && sudo make -j$(nproc) all install

ROS

For now only ROS Noetic is supported.

As an extra dependency install: sudo apt install ros-noetic-tf2-sensor-msgs. On your catkin workspace, clone the vdbfusion_ros:

sh cd catkin_ws/src/ git clone https://github.com/PRBonn/vdbfusion_ros.git catkin build

Usage

Configuration

1. Create a config file compatible with your dataset and desired tsdf integration parameters using this template configuration
2. The PointClouds should be a sensor_msgs/PointCloud2 message published on a custom topic name which needs to be specified in the config file
3. The Transforms should be either a tf2_msgs/TFMessage or a geometry_msgs/TransformStamped. See the template configuration for more details
4. The data can be published either through a rosbag file or directly from another ros node

Launch

sh roslaunch vdbfusion_ros vdbfusion.launch config_file_name:=<insert config file name here> path_to_rosbag_file:=<insert path to rosbag file here>

Save the VDB Grid and Extract Triangle Mesh

sh rosservice call /save_vdb_volume "path: '<insert filename and path to save the volume and mesh>'"

Dataset Examples

Download the dataset rosbag files from the respective links

TU Munich RGB-D SLAM Dataset and Benchmark - FR1DESK2

• Use the sample config file provided for this dataset

ETH Zurich ASL: Cow and Lady RGBD Dataset

• Use the sample config file provided for this dataset

KITTI Dataset

• Convert the dataset into a rosbag file using kitti2bag
• Use the sample config file provided for this dataset

Run the launch command providing config file and rosbag path corresponding to the dataset.

Use the rosservice to save the VDB volume and mesh

Citation

If you use this library for any academic work, please cite the original paper.

bibtex @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} }

Others ROS wrappers around VDBFusion

• vdbfusion_mapping: originated to solve transformation issues and with color support
• vdbfusionmappingdocker. Same as above, but running in a dockerized enviornment