CUDA SURF – A real-time implementation for SURF

CUDA SURF feature matching

Keypoint detection and matching is a basic computer vision task and a necessary ingredient for several applications, e.g., object recognition, structure from motion, panorama stitching. In this work we implement the popular SURF descriptor, an approximation of SIFT, on commodity graphics hardware and achieve real-time performance even for HD images. The code was originally developed as class project in the lecture “Programming Massively Parallel Processors” in the summer semester 2009. More infos and the source code are available at the CUDA SURF page, hosted at the Max Planck-Institut für Informatik.