A Real-Time Local Visual Feature for Omnidirectional Vision Based on FAST and CS-LBP

In this paper, a real-time local visual feature, namely FAST+CSLBP, is proposed for omnidirectional vision. It combines the advantages of two computationally simple operators by using Features from Accelerated Segment Test (FAST) as the feature detector, and Center-Symmetric Local Binary Patterns (CS-LBP) operator as the feature descriptor. The matching experiments of the panoramic images from the COLD database were performed to determine its best parameters, and to evaluate and compare its performance with SIFT. The experimental results show that our algorithm performs better, and features can be extracted in real-time. Therefore our local visual feature can be applied to the computer/robot vision tasks with high real-time requirements.