Expressing Relational and Temporal Knowledge in Visual Probabilistic Networks

TL;DR

This paper extends Bayesian networks to effectively model complex relational and temporal knowledge in high-level vision tasks, demonstrated through applications in endoscopy.

Contribution

It introduces a simple-structured extension of Bayesian networks for relational and temporal modeling in visual AI, enabling efficient probability propagation.

Findings

Extended networks handle complex relational and temporal knowledge.

Applied successfully to endoscopy domain.

Efficient probability propagation achieved.

Abstract

Bayesian networks have been used extensively in diagnostic tasks such as medicine, where they represent the dependency relations between a set of symptoms and a set of diseases. A criticism of this type of knowledge representation is that it is restricted to this kind of task, and that it cannot cope with the knowledge required in other artificial intelligence applications. For example, in computer vision, we require the ability to model complex knowledge, including temporal and relational factors. In this paper we extend Bayesian networks to model relational and temporal knowledge for high-level vision. These extended networks have a simple structure which permits us to propagate probability efficiently. We have applied them to the domain of endoscopy, illustrating how the general modelling principles can be used in specific cases.