抄録
In this paper we describe our novel work of using graphic processing unit (GPU) to improve a homographybased visual servo system. First we propose a GPU accelerated Efficient Second-order Minimization (GPU-ESM) algorithm. With so-called "GPU-ESM tracking algorithm", our system can provide a fast and stable homography solution, approximately 20 times faster than the CPU implementation. To enhance the system stability, we adopt a GPU based Scale Invariant Feature Transform (SIFT) algorithm to deal with those cases where GPU-ESM tracking algorithm performs poor, such as large image differences, occlusion and so on. In this paper, the combination of both GPU accelerated algorithms is described in detail. The effectiveness of our GPU accelerated system is evaluated with experimental data. Key optimization techniques in our GPU applications are presented as a reference for other researchers.
