ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question


asked 2012-08-13 06:04:57 -0500

dalspenguin gravatar image

updated 2014-01-28 17:13:19 -0500

ngrennan gravatar image

Hello ROS community,

How to use rgbd-slam with two cameras? I have read I run rgbd-slam from command line "roslaunch rgbdslam kinect+rgbdslam.launch" and it said: "cannot find kinect device". What should I put into command line in order to say rgbdslam reads frames(images) from two cameras, not from kinect device? Did anyone get success with it?

Also, I have tried to to use bagfiles instead of kinect device to get pre-saved frames(images), however did not get any success. Does anyone have example of using bagfiles?

My enviroment: Ubuntu 11.10 ROS electric 2 cameras

Thanks a lot, Alex

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2012-09-06 05:32:39 -0500

You need to compute a pointcloud from your stereo vision data, then have rgbdslam listen to the 2d image and the cloud. Set the parameters "wide_topic" and "wide_cloud_topic" to listen to the topics where the respective data is published.

The cloud must be dense, i.e. stored in a 2D grid where each point corresponds to one pixel and vice versa.

edit flag offensive delete link more

Question Tools

1 follower


Asked: 2012-08-13 06:04:57 -0500

Seen: 1,195 times

Last updated: Sep 06 '12