HAL : hal-00647995, version 1

arXiv : 1112.1187

DOI : 10.1109/TPAMI.2011.207

Fiche détaillée

Récupérer au format BibTeX - EndNote - TEI - RefWorks -

IEEE Transactions on Pattern Analysis and Machine Intelligence 34, 5 (2012) 930-942

Meaningful Matches in Stereovision

Neus Sabater 1, Andrés Almansa 2, Jean-Michel Morel 3

(05/2012)

This paper introduces a statistical method to decide whether two blocks in a pair of of images match reliably. The method ensures that the selected block matches are unlikely to have occurred "just by chance.'' The new approach is based on the definition of a simple but faithful statistical "background model" for image blocks learned from the image itself. A theorem guarantees that under this model not more than a fixed number of wrong matches occurs (on average) for the whole image. This fixed number (the number of false alarms) is the only method parameter. Furthermore, the number of false alarms associated with each match measures its reliability. This "a contrario" block-matching method, however, cannot rule out false matches due to the presence of periodic objects in the images. But it is successfully complemented by a parameterless "self-similarity threshold." Experimental evidence shows that the proposed method also detects occlusions and incoherent motions due to vehicles and pedestrians in non simultaneous stereo.

1 :	Centre de Mathématiques et de Leurs Applications - ENS Cachan (CMLA)
	École normale supérieure de Cachan - ENS Cachan
2 :	Laboratoire Traitement et Communication de l'Information [Paris] (LTCI)
	Télécom ParisTech – CNRS : UMR5141
3 :	Centre de Mathématiques et de Leurs Applications (CMLA)
	CNRS : UMR8536 – École normale supérieure de Cachan - ENS Cachan

Domaine

:

Informatique/Vision par ordinateur et reconnaissance de formes

Mots Clés : Stereo – block-matching – Number of False Alarms (NFA) – a contrario detection

hal-00647995, version 1

http://hal.archives-ouvertes.fr/hal-00647995

oai:hal.archives-ouvertes.fr:hal-00647995

Contributeur : Andrés Almansa <>

Soumis le : Lundi 5 Décembre 2011, 11:31:54

Dernière modification le : Samedi 24 Mars 2012, 01:46:11