Chatroom for the DUO Camera ROS Driver. Here is the place to get quick help and responses.
and i guess this is how it's done in openCV: http://opencv.jp/opencv-2svn_org/c/_images/math/dbe76ed3ef28b501c3933a74618c2c94889d38fa.png
i guess it should be possible since this (http://docs.carnegierobotics.com/SL/api.html) driver's docu says that the rational model is supported in ROS:
@gitunit can you check if this error occurs for you as well?
im in contact with DUO myself. they will provide some api changes to finish the correct loading of calibration info from device. currently i have a workaround but cant release the code since its very dirty as soon as i've cleaned it up, i will make a pull request. regarding the logs it is due to the current version, obviously it tries to load the on-device calibration but there is none because there is no checking before if there actually is a calibration on device. but if the user wants to use the old calibration files, just set the "use_DUO_calibration" to false. that should do it currently.
cheers
btw, i didn't release the driver for 1.4x as DUO still hadn't resolve some issues (rotated pointcloud etc.)
i havent had a chance to test it out myself
the latest duo driver should have fixed that issue but the api changed therefore i need some adjustments to benefit from the latest driver version
ok
Hey guys, I'm trying to take an openCV mat image and then convert it to a ros message. I'm using cv_bridge. An interesting thing I keep seeing is after I call " msg = cv_bridge::CvImage(header, "mono8", mat_image).toImageMsg();" if I call imshow, the image appears blank. if I call imshow right before it, i see the image. Why is that? Shouldnt this function not modify my mat image?
If anyone else was having config problems. FYI as of February DUO updated their SDK to make reading factory configuration data easier & directly compatible with OpenCV (and the ROS camera_info message) (https://duo3d.com/forums/viewthread/114/)
Since this is a robotics centered project I am going to try and ask here. I've been looking for a day now for a program which will stream video and add text. ffmpeg will add the text but I want the text to change. The text will have telemetry data from the sensors on my robot hense it needs to be dynamic. Think of it just like a heads up display. This would seem so common that there should be ready made stuff for it. Does anyone know where to look?
