rgbd_detection2d_3d
rgbd_detection2d_3d copied to clipboard
[ROS package] Project a 2D detection to 3D within an RGB-D camera
rgbd_detection2d_3d
Project a 2D detection to 3D within an RGB-D camera.
Notice
- The current version is tested with darknet_ros, that is, subscribing to
darknet_ros/bounding_boxes
(darknet_ros_msgs/BoundingBoxes) topic. If you need to subscribe to other types, please modify the code yourself. - The current published message type is vision_msgs/Detection3DArray. Similarly, please modify the code if other types are required.
- The current version requires that the point cloud must be ordered.
How to build
cd ~/catkin_ws/src/
git clone https://github.com/yzrobot/rgbd_detection2d_3d.git
cd ~/catkin_ws
catkin_make
Run
rosrun rgbd_detection2d_3d rgbd_detection2d_3d
Test environment
Ubuntu 20.04 LTS
ROS Noetic
Citation
If you are considering using this code, please reference the following:
@article{io22software,
author = {Iaroslav Okunevich and Vincent Hilaire and Stephane Galland and Olivier Lamotte and Liubov Shilova and Yassine Ruichek and Zhi Yan},
title = {Software-hardware Integration and Human-centered Benchmarking for Socially-compliant Robot Navigation},
journal = {CoRR},
volume = {abs/2210.15628},
year = {2022},
url = {http://arxiv.org/abs/2210.15628},
archivePrefix = {arXiv},
eprint = {2210.15628}
}