Hi, I am trying to combined with OpenCV to do image processing for frames of camera. Can I do such things in Qualcomm AR? We can set frameFormat from Qualcomm API, I am wondering can I use Frame::Image or Frame::mdata to get image data and then use OpenCV to load those images?
Thank you very much.
I don't know what pixel format OpenCV is expecting, but this should be possible. You can get the pixels for the camera image from the Image object. Start with the State object returned by the QCAR::Renderer::getInstance().begin() method. From there you can access the Frame object, and from the Frame object you can loop through the available Image objects and choose the one with the correct format. You can register for different pixel formats with the QCAR::setFrameFormat() method.
- Kim