An Evaluation of Volumetric Interest Points

Shape and texture are two fundamental properties of objects. The use of texture (i.e. 2D surface colouration), or appearance, for object detection, recognition and registration has been well studied, shape (in 3D) less so. However, the increasing prevalence of depth sensors and the diminishing returns to be had from appearance alone have seen a surge in shape-based methods. In this work we investigate the performance of several detectors of interest points in volumetric data, in terms of repeatability, number and nature of interest points. Such methods form the first step in many shape-based applications. Our detailed comparison, with both quantitative and qualitative measures on synthetic and real 3D data, both point-based and volumetric, enables readers to easily select a method suitable to their application.