Question about Azure Kinect Depth information about ros_openpose HOT 2 CLOSED

That worked thank you so much for your help!!!

from ros_openpose.

@jmberd02

Thanks a lot for creating this issue. I hope you are running the following command:

roslaunch ros_openpose run.launch camera:=azurekinect

I definitely get 2d output in the console, however, I do not see a Z entry

You are right. That's because we have set to print only pixel information. Please see the code below:

You can notice that bodyPart.pixel is used here to print in the console. It denotes 2D information.

is there some way to get that information?

Yes. There is. Please edit bodyPart.pixel to bodyPart.point as shown below:

text = [bodyPart.point for person in msg.persons for bodyPart in person.bodyParts]
rospy.loginfo('%s\n' % text)

Extra Information

Just in case, if you are running the synchronous API version, you must be using it in the following way:

roslaunch ros_openpose run.launch camera:=azurekinect synchronous:=true py_openpose_path:=absolute_path_to_py_openpose

However, recently a bug has been discovered in the synchronous API version. Please see here for more information.

from ros_openpose.