Speeding up Haartraining with AMD GPU and OpenCL
Hello. What are the steps required for accelerating haartraining with OpenCL GPU?
I'm using openCV 2.4.5. I managed to build with OpenCL support, I have a AMD Radeon GPU but I can't figure out how to make the haartrainer use the GPU. My OpenCL detects the GPU correctly. I use the opencv_haartraining binary. Do I need to modify any source code for this to work ?