Modelling Distributed Shape Priors by Gibbs Random Fields of Second Order

TL;DR

This paper explores the use of second order Gibbs Random Fields to model and recognize complex shapes by representing simple shapes and their spatial relations, enabling effective shape prior modeling.

Contribution

It demonstrates that second order GRFs are sufficiently expressive for modeling complex shapes and their spatial compositions, advancing shape recognition methods.

Findings

Second order GRFs can model simple shapes effectively.

They can also encode spatial relations between shapes.

This approach enables recognition of complex shape compositions.

Abstract

We analyse the potential of Gibbs Random Fields for shape prior modelling. We show that the expressive power of second order GRFs is already sufficient to express simple shapes and spatial relations between them simultaneously. This allows to model and recognise complex shapes as spatial compositions of simpler parts.