Unable to convert '32FC1' image for display: 'cv_bridge.cvtColorForDisplay()
Hello~ all
I just followed jetsonhacks tutorial to install ROS-kinetic on my Nvidia TX1 successfully. Then git clone git clone link text to install Microsoft Kinect V1 Driver. The steps is as follows:
$mkdir ~/kinectdriver
$cd ~/kinectdriver
$git clone https://github.com/avin2/SensorKinect/tree/master
$cd SensorKinect/Bin/
$tar xvjf SensorKinect093-Bin-Linux-x64-v5.1.2.1.tar.bz2
$cd Sensor-Bin-Linux-x64-v5.1.2.1/
$sudo ./install.sh
To test Kinect, I do these on different Terminal to show Depth Image:
- roscore
- roslaunch openni_launch openni.launch
- rosrun image_view image_view image:=/camera/rgb/image_color (<- RGB image shows successfully )
- rosrun image_view image_view image:=/camera/depth/image (<- Failed to show depth)
And an error occured with this info:
[ERROR] [1480065718.281091328]: Unable to convert '32FC1' image for display: 'cv_bridge.cvtColorForDisplay() while trying to convert image from '32FC1' to 'bgr8' an exception was thrown ([32FC1] is not a color format. but [bgr8] is. The conversion does not make sense)'
It looks like some problems of pixel format conversation. I google this and know the solution is to modify code. But the method is not clean... Anyone knows how to fix it?
Thank you!!
the solution link is link text
Using Indigo, i have the same problem when using the example from "ROS By Example" volume 1 chapter 10. I use: rosrun image_view image_view image:=/camera/depth/image_rect, and get the SAME error.
The strange thing is that it DID WORK a few months ago. What has changed since then?
Forgot to mention. I use a regular PC, no nvidia stuff.
@Sietse I got the same problem, have you found any solution? By the way, in Chapter 10.11, I can see the Point Clouds by following the tutorial's instructions.
@Sietes , I saw the depth vision now, see here:another answer