| Summary: | In this work we propose new surface descriptors for matching and recognition of three-dimensional rigid objects by encoding the fluctuation of the surface and the variation of its normal around an oriented surface point. The surface of the object is encoded into three features vectors as the surface signatures on each point of the surface, and then the collection of signatures is used to match and recognize the object. The signatures encode the curvature, symmetry, and convexity of the surface around an oriented point. The new descriptors are robust to noise, sampling density, scale, rotation, clutter, and occlusion. In this work we use Computer Aided Design (CAD) to create models and test objects represented by triangular meshes. === Applied Science, Faculty of === Electrical and Computer Engineering, Department of === Graduate
|
|---|
