MDL, Collineations and the Fundamental Matrix

Steve Maybank (1) , Peter Sturm (1)

1 Computer Science Department

Scene geometry can be inferred from point correspondences between two images. The inference process includes the selection of a model. Four models are considered: background (or null), collineation, affine fundamental matrix and fundamental matrix. It is shown how Minimum Description Length (MDL) can be used to compare the different models. The main result is that there is little reason for preferring the fundamental matrix model over the collineation model, even when the former is the "true" model.

Domaines: Synthèse d'image et réalité virtuelle [cs.GR]

Steve Maybank, Peter Sturm. MDL, Collineations and the Fundamental Matrix. The 10th British Machine Vision Conference (BMVC '99), Sep 1999, Nottingham, United Kingdom. pp.53-62. ⟨inria-00525677⟩