ACCURATE DENSE STEREO MATCHING FOR ROAD SCENES - Université de Bourgogne Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

ACCURATE DENSE STEREO MATCHING FOR ROAD SCENES

Oussama Zeglazi
  • Fonction : Auteur correspondant
  • PersonId : 1017310

Connectez-vous pour contacter l'auteur
Mohammed Rziza
  • Fonction : Auteur
  • PersonId : 997751
Aouatif Amine
  • Fonction : Auteur

Résumé

Stereo matching task is the core of applications linked to the intelligent vehicles. In this paper, we present a new variant function of the Census Transform (CT) which is more robust against radiometric changes in real road scenes. We demonstrate that the proposed cost function outperforms the conventional cost functions using the KITTI benchmark. The cost aggregation method is also updated for taking into account the edge information. This enables to improve significantly the aggregated costs especially within homogenous regions. The Winner-Takes-All (WTA) strategy is used to compute disparity values. To further eliminate the remainder matching ambiguities , a post-processing step is performed. Experiments were conducted on the new Middlebury 2 dataset, as well as on the real road traffic scenes of the KITTI database. Obtained disparity results have demonstrated that the proposed method is promising.
Fichier principal
Vignette du fichier
ICIP2017.pdf (2.8 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01592090 , version 1 (22-09-2017)

Identifiants

  • HAL Id : hal-01592090 , version 1

Citer

Oussama Zeglazi, Mohammed Rziza, Aouatif Amine, Cédric Demonceaux. ACCURATE DENSE STEREO MATCHING FOR ROAD SCENES. IEEE International Conference on Image Processing, Sep 2017, Beijing, China. ⟨hal-01592090⟩
404 Consultations
226 Téléchargements

Partager

Gmail Facebook X LinkedIn More