This paper presents GpuCV, an open source multi-platform library for easily developing GPU-accelerated image processing and Computer Vision operators and applications. It is meant for computer vision scientist not familiar with GPU technologies. It is designed to be compatible with Intel's OpenCV library by offering GPU-accelerated operators that can be integrated into native OpenCV applications. The GpuCV framework transparently manages hardware capabilities, data synchronization, activation of low level GLSL and CUDA programs, on-the-fly benchmarking and switching to the most efficient implementation and finally offers a set of image processing operators with GPU acceleration available.

Recently researchers have shown that it is possible to use GPU hardware for image processing and computer vision algorithms. We have been exploring how to use GPU hardware to improve marker- based tracking for AR Applications. In this paper we describe our findings and explored issues in the context of a standard fiducial tracking pipeline. We demonstrate the implementation of a template matching process on the GPU and the performance improvement gained in comparison to a traditional CPU implementation.

This paper presents a state of the art report on using graphics hardware for image processing and computer vision. Then we describe GPUCV, an open library for easily developing GPU accelerated image processing and analysis operators and applications