| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
I have extracted an object of interest using Euclidean Cluster Extraction in PCL. I would now like to extract the corresponding object in a 2D image that has been saved simultaneously. Are there any tools available in ros/pcl that can be used to do this?
PCL has tools to generate an image from the point cloud directly (basic or range image), however not for dealing with two parallel objects. The above links will work if you're willing to take a hit in quality, but if you want to work with the original RGB image, you'll have to do some manual calculations.
You could project the pointcloud onto a plane representing the rectified* image, and use the points to mark which pixels you want to 'keep'. Or just project the bounding box and.
*: I am not a vision expert, so this may not be rigorous
Asked: 2014-09-26 08:57:38 -0500
Seen: 572 times
Last updated: Sep 26 '14
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
