Towards a General Vision System Based on Symbol-Relation Grammars and Bayesian Networks

Abstract. A novel approach to create a general vision system is presented. The proposed method is based on a visual grammar representation which is transformed to a Bayesian network which is used for object recognition. We use a symbol-relational grammar for a hierarchical description of objects, incorporating spatial relations. The structure of a Bayesian network is obtained automatically form the grammar, and its parameters are learned from examples. The method is illustrated with two examples for face recognition.

• Type: Journal
• Year: 2011
• Where: AGI
• Authors: Elias Ruiz, Augusto Meléndez, Luis Enrique Sucar
• Keywords: AGI 2011, Artificial Intelligence, Bayesian Network, Relational Grammar, Visual Grammar