Scene understanding by rule evaluation
W.F. Blaschof, Terry Caelli · IEEE Transactions on Pattern Analysis and Machine Intelligence · 1997
We consider how machine learning can be used to help solve the problem of identifying objects or structures composed of parts in complex scenes. We first discuss a conditional rule generation technique that is designed to describe structures using part attributes and their relations. We then show how the resultant rules can be used for region labeling and examine constraint propagation techniques for improving rule-based object classification.