Learning to Detect Ground Control Points for Improving the Accuracy of Stereo Matching

Aristotle Spyropoulos 1 Nikos Komodakis 2, 3 Philippos Mordohai 1, *
* Auteur correspondant
2 IMAGINE [Marne-la-Vallée]
LIGM - Laboratoire d'Informatique Gaspard-Monge, CSTB - Centre Scientifique et Technique du Bâtiment, ENPC - École des Ponts ParisTech
Abstract : While machine learning has been instrumental to the ongoing progress in most areas of computer vision, it has not been applied to the problem of stereo matching with similar frequency or success. We present a supervised learning approach for predicting the correctness of stereo matches based on a random forest and a set of features that capture various forms of information about each pixel.We show highly competitive results in predicting the correctness of matches and in confidence estimation, which allows us to rank pixels according to the reliability of their assigned disparities. Moreover, we show how these confidence values can be used to improve the accuracy of disparity maps by integrating them with an MRF-based stereo algorithm. This is an important distinction from current literature that has mainly focused on sparsification by removing potentially erroneous disparities to generate quasi-dense disparity maps.
Type de document :
Communication dans un congrès
Computer Vision and Pattern Recognition 2014, Jun 2014, Columbus, Ohio, United States. 2014, 〈10.1109/CVPR.2014.210〉
Liste complète des métadonnées

Littérature citée [26 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01246487
Contributeur : Nikos Komodakis <>
Soumis le : vendredi 18 décembre 2015 - 15:55:26
Dernière modification le : jeudi 5 juillet 2018 - 14:29:16
Document(s) archivé(s) le : samedi 19 mars 2016 - 14:20:13

Fichier

cvpr2014.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Aristotle Spyropoulos, Nikos Komodakis, Philippos Mordohai. Learning to Detect Ground Control Points for Improving the Accuracy of Stereo Matching. Computer Vision and Pattern Recognition 2014, Jun 2014, Columbus, Ohio, United States. 2014, 〈10.1109/CVPR.2014.210〉. 〈hal-01246487〉

Partager

Métriques

Consultations de la notice

327

Téléchargements de fichiers

88