Two novel real-time local visual features for omnidirectional vision

Two novel real-time local visual features, namely FAST+LBP and FAST+CSLBP, are proposed in this paper for omnidirectional vision. They combine the advantages of two computationally simple operators by using FAST as the feature detector, and LBP and CS-LBP operators as feature descriptors. The matching experiments of the panoramic images from the COLD database were performed to determine their optimal parameters, and to evaluate and compare their performance with SIFT. The experimental results show that our algorithms perform better, and features can be extracted in real-time. Therefore our local visual features can be applied to those computer/robot vision tasks with high real-time requirements.