Using spin images for efficient object recognition in cluttered 3D scenes：（使用旋转图像在凌乱的3 d场景有效的目标识别）.pdfVIP

Using Spin-Images for Efficient Object Recognition in Cluttered 3-D Scenes Andrew E. Johnson Martial Hebert Jet Propulsion Laboratory The Robotics Institute Mail Stop 107-102 Carnegie Mellon University 4800 Oak Grove Drive 5000 Forbes Avenue Pasadena, CA 91109 Pittsburgh, PA 15213 aej@robotics.jpl.nasa.gov hebert@ri.cmu.edu Submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence submitted: July 1998 Abstract We present a 3-D shape-based object recognition system for simultaneous recognition of multiple objects in scenes containing clutter and occlusion. Recognition is based on matching surfaces by matching points using the spin-image representation. The spin-image is a data level shape descrip- tor that is used to match surfaces represented as surface meshes. We present a compression scheme for spin-images that results in efficient multiple object recognition which we verify with results showing the simultaneous recognition of multiple objects from a library of 20 models. Fur- thermore, we demonstrate the robust performance of recognition in the presence of clutter and oc- clusion through analysis of recognition trials on 100 scenes. This research was performed at Carnegie Mellon University and was supported by the US Depart- ment of Energy under contract DE-AC21-92MC29104. 1 Introduction Surface matching is a technique from 3-D computer vision that has many applications in the area of robotics and automation. Through surface matching, an object can be recognized in a scene by comparing a sensed surface to an object surface stored in memory. When the object surface is matched to the scene surface, an associat