Joint hierarchical learning for efficient multi-class object detection

Abstract : In addition to multi-class classification, the multi-class object detection task consists further in classifying a dominating background label. In this work, we present a novel approach where relevant classes are ranked higher and background labels are rejected. To this end, we arrange the classes into a tree structure where the classifiers are trained in a joint framework combining ranking and classification constraints. Our convex problem formulation naturally allows to apply a tree traversal algorithm that searches for the best class label and progressively rejects background labels. We evaluate our approach on the PASCAL VOC 2007 dataset and show a considerable speed-up of the detection time with increased detection performance.
Type de document :
Communication dans un congrès
Applications of Computer Vision (WACV), 2014 IEEE Winter Conference on, Mar 2014, Steamboat Springs, United States. pp.261--268, 2014
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01296239
Contributeur : Thierry Chateau <>
Soumis le : jeudi 31 mars 2016 - 17:11:24
Dernière modification le : lundi 24 septembre 2018 - 11:34:03

Identifiants

  • HAL Id : hal-01296239, version 1

Collections

Citation

Hamidreza Odabai Fard, Mohamed Chaouch, Quoc-Cuong Pham, Antoine Vacavant, Thierry Chateau. Joint hierarchical learning for efficient multi-class object detection. Applications of Computer Vision (WACV), 2014 IEEE Winter Conference on, Mar 2014, Steamboat Springs, United States. pp.261--268, 2014. 〈hal-01296239〉

Partager

Métriques

Consultations de la notice

141