Hessian Interest Points on GPU

This paper is about interest point detection and GPU programming. We take a popular GPGPU implementation of SIFT – the de-facto standard in fast interest point detectors – SiftGPU and implement modifications that according to recent research result in better performance in terms of repeatability of the detected points. The interest points found at local extrema of the Difference of Gaussians (DoG) function in the original SIFT are replaced by the local extrema of determinant of Hessian matrix of the intensity function. Experimentally we show that the GPU implementation of Hessian-based detector (i) surpasses in repeatability the original DoG-based implementation, (ii) gives result very close to those of a reference CPU implementation, and (iii) is significantly faster than the CPU implementation. We show what speedup is achieved for different image sizes and provide analysis of computational cost of individual steps of the algorithm. The source code is publicly available.