Multi-scale feature extraction for 3D models using local surface curvature

In this paper, we present a method for extracting salient local features from 3D models using surface curvature which has application to 3D object recognition. In the developed technique, the amount of curvature at a point is specified by a positive number known as the curvedness. This value is invariant to rotation as well as translation. A local description of the surface is generated by fitting a surface to the neighbourhood of a keypoint and estimating its curvedness at multiple scales. From this surface, points corresponding to local maxima and minima of curvedness are selected as suitable features and a confidence measure of each keypoint is also calculated based on the deviation of its curvedness from the neighbouring values. Experimental results on a different number of models are shown to demonstrate the effectiveness and robustness of our approach