Based on visual SLAM for markerless augmented reality,an optimized markerless augmented reality method is pro-posed.Before plane fitting,a point cloud splicing module based on point cloud overlap rate is added to make the transition of contin-uous frame fitting plane smoother.In order to improve the plane fitting accuracy of the plane fitting stage,the RANSAC algorithm with adaptive threshold is used on the basis of the principal component analysis method to eliminate outliers and gross errors in the point cloud data.The proposed method is a method with higher fitting accuracy and stability,which can be verified by the test re-sults.
维普
万方数据