HI all,
I am working on optimizing a video processing system using CUDa… As of now we capture the image from the Camera using the camera’s SDK functions and transfer the image to the GPU to do the processing and return back to PC. I noticed the camera can capture in DIrectDraw mode where the image is stored in the VGA memory directly using Overlay buffer.
The question is… How to use CUDA to read this memory. ? I need to calculate its mean to start with… so any ideas on how to read it pixel wise will be very helpful
I hope someone has an answer to this…
Thanks
Ash