preetham.hegde's profile - activity

Hi, Does anybody know how kinect data is processed in ROS. as i know it has a RGB Data and a Depth Data usually stored in a .xed file. Anybody knows how its processed in ROS??

If its split into 2 files or seperate streams etc?? Does OpenNi do it for ROS??

Hello all, I am new to ROS and i am trying to use the RGBDSLAM with ROSbag files i.e i want to first run the rosbag record and make a rosbag file and then run RGBD SLAM such that it will read this rosbag file and give me a map.

I found another post on the similar issue and used the launch file he used i.e

<launch>
  <node pkg="rgbdslam" type="rgbdslam" name="rgbdslam" cwd="node" required="false" output="log" > 
    <!-- see rgbslam_sample_config.launch for all available parameters and their default values -->
      <param name="config/topic_image_mono"              value="/camera/rgb/image_color"/>

<param name="config/topic_image_depth"             value="/camera/depth_registered/image"/>
<param name="config/topic_points"                  value=""/> <!--if empty, poincloud will be reconstructed from image and depth -->
<param name="config/camera_info_topic"             value="/camera/rgb/camera_info"/>
<param name="config/bagfile_name"                  value="/home/preetham/Desktop/filename.bag"/>
<param name="config/start_paused"                  value="true"/>

      <param name="config/wide_topic"                    value=""/>;
      <param name="config/wide_cloud_topic"              value=""/>;
      <param name="config/drop_async_frames"             value="true"/> <!-- Check association of depth and visual image, reject if not in sync -->
      <param name="config/feature_detector_type"         value="SURF"/><!-- If SIFTGPU is enabled in CMakeLists.txt, use SURF here -->
      <param name="config/feature_extractor_type"        value="SURF"/><!-- If SIFTGPU is enabled in CMakeLists.txt, use SURF here -->
      <param name="config/matcher_type"                  value="FLANN"/> <!-- FLANN (not avail for ORB features), SIFTGPU (only for SIFTGPU detector)  or BRUTEFORCE-->
      <param name="config/max_keypoints"                 value="2000"/><!-- Extract no more than this many keypoints (not honored by SIFTGPU)-->
      <param name="config/min_keypoints"                 value="1600"/><!-- Extract no less than this many ... -->
      <param name="config/optimizer_skip_step"           value="5"/><!-- optimize every n-th frame -->
      <param name="config/store_pointclouds"             value="true"/> <!-- if, e.g., only trajectory is required, setting this to false saves lots of memory -->
      <param name="config/individual_cloud_out_topic"    value="/rgbdslam/batch_clouds"/>;
      <param name="/config/use_gui" value="true"/>
    <!-- see rgbslam_sample_config.launch for all available parameters and their default values -->

<remap from="/camera/depth/camera_info" to="/camera/depth_registered/camera_info" />

  </node>
</launch>

And when i launch RGBDSLAM using this launch file it throws an error.

The following is the rosbag command i used and it records a rosbag file of size 300 Mb for 10 seconds which i feel is correct.

rosbag record -o /home/preetham/Desktop/filename_prefix /tf /camera/rgb/image_color /camera/rgb/camera_info /camera/depth_registered/image /camera/depth/camera_info

ould somebody help me out??

Thanks @Daniel Stonier. I will try it out and get back with an answer as soon as possible.

Hello Guys, I recently used RGBD Slam for my project from my Ubuntu Machine and it did work great. Really Cool Stuff. However i have both the Kinect for XBox and Kinect for Windows. I want to use the Kinect for Windows as it has a Near Mode capability and i believe the Maps could be better with it. So i plan to install WinRos and run ROSBAG from it and bag all the required files and then transfer the files to Linux and run it alongside RGBDSLAM. Is this Possible??

If its possible could somebody guide me some steps on how to use WinRos and ROSBAG on it. I don't see proper documentation for it.

Hello All, I am running RGBDSLAM on my Laptop and it loads the GUI and shows me the monochrome image and depth image and when i hit Space or Enter to start the Point CLoud Processing it fails and the terminal shows follwing error.

¨Orientation simplified on this hardware Descriptor ignored on this hardware OpenCV Error: Unsupported format or combination of formats (Unsupported combination of input and output array formats) in reduce, file /tmp/buildd/ros-fuerte-opencv2-2.4.2- 1precise-20130312-1308/modules/core/src/matrix.cpp, line 2242 terminate called after throwing an instance of 'cv::Exception' what(): /tmp/buildd/ros-fuerte-opencv2-2.4.2-1precise-20130312-1308/modules/core/src/matrix.cpp:2242: error: (-210) Unsupported combination of input and output array formats in function reduce

================================================================================REQUIRED process [rgbdslam-23] has died! process has died [pid 14077, exit code -6, cmd /home/preetham/ros/rgbdslam_freiburg/rgbdslam/bin/rgbdslam __name:=rgbdslam __log:=/home/preetham/.ros/log/7fbba3e8-a59f-11e2-9891-844bf5768cdb/rgbdslam-23.log]. log file: /home/preetham/.ros/log/7fbba3e8-a59f-11e2-9891-844bf5768cdb/rgbdslam-23*.log Initiating shutdown!¨

Anybody can suggest me what the problem is and how i can go about it??

@Merdad: Did u get it to work finally. I am having the same issue and still couldn't figure out the issue. http://answers.ros.org/question/57412/cant-see-kinect-data/

Its not on VMware. This is on my HP Envy Notebook. Its a Dual Boot Ubuntu 12.04 and WIndows 8. And its USB 2.0.

I also tried opening RVIZ and it failed and even trying to access a single image did not show up anything. Help Needed Please.

@Felix Endres: I had tried with the binaries and they had failed. I removed ros totally and followed the steps that were mentioned at the link you posted and this time RGBD Installed perfect. The GUI Opened but i could see nothing. I pressed space and enter ,it said processing but nothing showed up.

OpenNI :- https://github.com/OpenNI/OpenNI/tree/unstable Sensor Kinect:- https://github.com/avin2/SensorKinect NITE:- Nite 1.5.2.21 for WIn64. Are these the right softwares?? I installed Ubuntu-> Then ROS-> Then OpenNI -> then Sensor Kinect-> then Nite.

@Stephane.M & @Felix Endres:- I did all the installs and it still is the same. I am guessing i am installing the wrong software. Can anybody guide me to the right link for the softwares. Here is the link to what i used:

@Felix Endres: I would like to do a fresh Install of Ubuntu and then Kinect drivers & ROS. Could you give me a quick description on how to do it. Or a link of any good article you know on how to do it.

@Felix Endres: I followed this guide to install the drivers. Let me know if this was the right one or was there a different one?? Link: http://mitchtech.net/ubuntu-kinect-openni-primesense/

@Stephane.M : Yeah ive tried that. When i launch RViz i get a display_config not found error. But then i click OK and proceed and i configure as mentioned in the Rviz Tutorials in ROS. But they simply show a blank window. nothing displayed in it.

@Felix Endres: Is this the output you were asking for?? Or is it a differnt one?? let me know which one u need exactly and i will post it. Sorry but i am new to ROS and this community and am getting things slowly.

And on running rxgraph i get a graph in which all paths are pointing an arrow towards/rosout.

@Felix Endres : On running roswtf i get 17 Errors "Communication with /camera/depth/metric/ failed

Even subscribing to the rectified topics just says subscribed and nothing more happens.

And when i run a rostopic it just says subscribed to the /camera/depth_registered/points and keeps blinking.

Well when i launch the openni.launch it starts as usual except for the errors ¨Tried to advertise a service that is already advertised in this node [/camera/depth_registered/image_rect_raw/compressedDepth/set_parameters]¨

Hi, I am working on this project that involves using the Kinect for XBOX 360S with ROS.

I did all the steps mentioned in the ROS Tutorials to have Openni Installed and the Prime sense and other drivers. and when i go to Openni samples i see a output. But in ROS i do a roscore and in another terminal do a roslaunch openni_launch openni.launch. And it loads with the regular calibration warnings and service already registered errors. Then in another terminal i open Rviz which gives a error /.rviz/display_config does not exist. And even though i accept the error and go ahead i see a black window which shows no output ,even if i do all tasks mentioned at the RVIZ Tutorials. Also i tried running "rosrun image_view image_view image:=/camera/rgb/image_color" and it shows up a blank window with no output. How do i resolve this and get ros to show my kinect data??

I need to run RGBDSLAM and use this kinect later.

I am on Ubuntu 12.04 and ROS-Fuerte.