high performance computing on graphics processing units: hgpu.org

Pedestrian detection has gained a lot of prominence during the last few years. Besides the fact that it is one of the hardest tasks within computer vision, it involves huge computational costs. Obtaining acceptable real-time performance, measured in frames per second (fps), for the most advanced algorithms is nowadays a hard challenge. In this work, we propose a GPU implementation of a well-known pedestrian detection system (i.e., HOGLBP-SVM) specially designed for the Tegra X1 embedded GPU. It includes LBP and HOG as feature descriptors and SVM as classifiers. We introduce significant algorithmic adjustments and optimizations to adapt the problem to the NVIDIA GPU architecture without sacrificing accuracy. The aim of this work is to offer a real-time system providing reliable results.