3D object recognition based on a viewpoint-analysis

Abstract An approach for the recognition of 3D objects is proposed. The approach is based on view-analysis and compilation of those aspects related to each of the objects being recognized, i.e. the expectations of appearance of the objects related to each viewpoint are generated and analyzed so as to discard viewpoints related to similar surfaces. For each object a 3D multiview description is introduced; these descriptions are integrated in a tree-structure data in order to simplify matching of the unknown object against every possible model.

[1]  Berthold K. P. Horn.  Robot vision , 1986, MIT electrical engineering and computer science series.

[2]  Nicola Veneziani,et al.  A two-pass filling algorithm for raster graphics , 1982, Comput. Graph. Image Process..

[3]  Rodney A. Brooks,et al.  Symbolic Reasoning Among 3-D Models and 2-D Images , 1981, Artif. Intell..

[4]  I. K. Sethi.  A general scheme for discontinuity detection , 1985, Pattern Recognit. Lett..

[5]  Yoshiaki Shirai,et al.  Object Recognition Using Three-Dimensional Information , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Nils J. Nilsson,et al.  Principles of Artificial Intelligence , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Charles R. Dyer,et al.  3-D multiview object representations for model-based object recognition , 1987, Pattern Recognit..

[8]  Nicola Ancona,et al.  A model-based 3-D vision system for bin-picking , 1988 .