| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
I'd like to save raw kinect data with minimal computer overhead to achieve maximal framerate on a computer that may not have enough performance for much else, and then take the saved data and do the depth-rgb alignment and conversion to point cloud non-real-time.
What tools and commands should I use?
I've found some discussion of the same question here: http://kinect-with-ros.976505.n3.nabble.com/Ros-kinect-Point-cloud-projection-in-a-second-pass-for-CPU-limited-systems-td2497149.html#a2756101
When I get a chance to try it out I'd like to produce a tutorial with every command line documented, unless someone else beats me to it.
dynamic reconfigure lets you turn off registration of the two images. Then you can use rosbag to save the rgb and depth image topics for later processing. Or you could edit openni_node.cpp in the openni_camera package to store the images instead of publishing, but this is far more involved.
Asked: 2011-03-25 03:19:14 -0600
Seen: 1,549 times
Last updated: Mar 31 '11
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.
