Model-Based 3D Object Recognition Using Bayesian Indexing

This research features the rapid recognition of three-dimensional objects, focusing on efficient indexing. A major concern in practical vision systems is how to retrieve the best matched models without exploring all possible object matches. We have employed a Bayesian framework to achieve efficient indexing of model objects. A decision-theoretic measure of the discriminatory power of a feature for a model object is defined in terms of posterior probability. Domain-specific knowledge compiled off-line from CAD model data is used in order to estimate posterior probabilities that define the discriminatory power of features for model objects. In order to speed up the indexing or selection of correct objects, we generate and verify the object hypotheses for features detected in a scene in the order of the discriminatory power of these features for model objects. Based on the principles described above, we have implemented a working prototype vision system using a feature structure called an LSG (local surface group) for generating object hypotheses. Our object recognition system can employ a wide class of features for generation of object hypotheses. In order to verify an object hypothesis, we estimate the view of the hypothesized model object and render the model object for the computed view. The object hypothesis is then verified by finding additional features in the scene that match those present in the rendered image. Experimental results on synthetic and real range images show the effectiveness of the indexing scheme.