What, Where & How Many? Combining Object Detectors and CRFs

Abstract : Computer vision algorithms for individual tasks such as object recognition, detection and segmentation have shown impressive results in the recent past. The next challenge is to integrate all these algorithms and address the problem of scene understanding. This paper is a step towards this goal. We present a probabilistic framework for reasoning about regions, objects, and their attributes such as object class, location, and spatial extent. Our model is a Conditional Random Field defined on pixels, segments and objects. We define a global energy function for the model, which combines results from sliding window detectors, and low-level pixel-based unary and pairwise relations. One of our primary contributions is to show that this energy function can be solved efficiently. Experimental results show that our model achieves significant improvement over the baseline methods on CamVid and PASCAL VOC datasets.
Type de document :
Communication dans un congrès
ECCV - European Conference on Computer Vision, Sep 2010, Heraklion, Crete, Greece. Springer, 〈10.1007/978-3-642-15561-1_31〉
Liste complète des métadonnées

Littérature citée [33 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01216730
Contributeur : Karteek Alahari <>
Soumis le : vendredi 16 octobre 2015 - 18:09:07
Dernière modification le : lundi 28 mai 2018 - 15:10:06
Document(s) archivé(s) le : lundi 18 janvier 2016 - 06:07:19

Fichier

ladicky10.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Lubor Ladicky, Paul Sturgess, Karteek Alahari, Chris Russell, Philip H. S. Torr. What, Where & How Many? Combining Object Detectors and CRFs. ECCV - European Conference on Computer Vision, Sep 2010, Heraklion, Crete, Greece. Springer, 〈10.1007/978-3-642-15561-1_31〉. 〈hal-01216730〉

Partager

Métriques

Consultations de la notice

26

Téléchargements de fichiers

202