Probabilistic indexing: a new method of indexing 3D model data from 2D image data

Recent research has indicated that indexing is a promising approach to fast model-based object recognition because it allows most of the possible matches between image point groups and model point groups to be quickly eliminated from consideration. Current indexing systems for the problem of recognizing general 3-D objects from single 2-D images require groups of four points to generate a key into the index table and each model group requires many entries in the table. The author presents a system that is capable of indexing using groups of three points by taking advantage of the probabilistic peaking effect. Each model group need only be represented at one point in the index table. The ability to index using groups of three points means that there are many fewer model groups and image groups to consider, but to be able to index using groups of three points, false negatives matches must be allowed. These false negatives can be withstood by examining information from multiple groups. Results are given on real and synthetic data