Improving recognition of complex aerial scenes using a deep weakly supervised learning paradigm - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année :

Improving recognition of complex aerial scenes using a deep weakly supervised learning paradigm

(1) , (1)
1
Nikos Komodakis
  • Fonction : Auteur
  • PersonId : 945099

Résumé

Categorizing highly complex aerial scenes is quite strenuous due to the presence of detailed information with large number of distinctive objects. Recognition happens by first deriving a joint relationship within all these distinguishing objects, distilling finally to some meaningful knowledge that is subsequently employed to label the scene. However, something intriguing is whether all this captured information is actually relevant to classify such a complex scene ? What if some objects just create uncertainty with respect to the target label, thereby causing ambiguity in the decision making ? In this paper, we investigate these questions and analyze as to which regions in an aerial scene are the most relevant and which are inhibiting in determining the image label accurately. However, for such Aerial Scene Classification (ASC) task, employing supervised knowledge of experts to annotate these discriminative regions is quite costly and laborious; especially when the dataset is huge. To this end, we propose a Deep Weakly Supervised Learning (DWSL) technique. Our classification-trained Convolutional Neural Network (CNN) learns to identify discriminative region localizations in an aerial scene solely by utilizing image labels. Using the DWSL model, we significantly improve the recognition accuracies of highly complex scenes, thus validating that extra information causes uncertainty in decision making. Moreover, our DWSL methodology can also be leveraged as a novel tool for concrete visualization of the most informative regions relevant to accurately classify an aerial scene. Lastly our proposed framework yields state-of-the-art performance on existing ASC datasets.
Fichier principal
Vignette du fichier
Manuscript_pdf.pdf (3.63 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01856457 , version 1 (11-08-2018)

Identifiants

  • HAL Id : hal-01856457 , version 1

Citer

Praveer Singh, Nikos Komodakis. Improving recognition of complex aerial scenes using a deep weakly supervised learning paradigm. 2018. ⟨hal-01856457⟩
84 Consultations
175 Téléchargements

Partager

Gmail Facebook Twitter LinkedIn More