LocNet: Improving Localization Accuracy for Object Detection

Abstract : We propose a novel object localization methodology with the purpose of boosting the localization accuracy of state-of-the-art object detection systems. Our model, given a search region, aims at returning the bounding box of an object of interest inside this region. To accomplish its goal, it relies on assigning conditional probabilities to each row and column of this region, where these probabilities provide useful information regarding the location of the boundaries of the object inside the search region and allow the accurate inference of the object bounding box under a simple probabilistic framework. For implementing our localization model, we make use of a convolutional neural network architecture that is properly adapted for this task, called LocNet. We show experimentally that LocNet achieves a very significant improvement on the mAP for high IoU thresholds on PASCAL VOC2007 test set and that it can be very easily coupled with recent state-of-the-art object detection systems, helping them to boost their performance. Furthermore, it sets a new state-of-the-art on PASCAL VOC2012 test set achieving mAP of 74.8%. Finally, we demonstrate that our detection approach can achieve high detection accuracy even when it is given as input a set of sliding windows, thus proving that it is independent of bounding box proposal methods.
Type de document :
Rapport
[Technical Report] Ecole des Ponts ParisTech. 2015
Liste complète des métadonnées

Littérature citée [35 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01245707
Contributeur : Spyros Gidaris <>
Soumis le : jeudi 17 décembre 2015 - 15:18:13
Dernière modification le : jeudi 11 janvier 2018 - 06:20:22
Document(s) archivé(s) le : vendredi 18 mars 2016 - 14:00:28

Fichier

1511.07763v1.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01245707, version 1
  • ARXIV : 1511.07763

Citation

Spyros Gidaris, Nikos Komodakis. LocNet: Improving Localization Accuracy for Object Detection. [Technical Report] Ecole des Ponts ParisTech. 2015. 〈hal-01245707〉

Partager

Métriques

Consultations de la notice

361

Téléchargements de fichiers

1123