Centre de documentation
scientifique

Fine-grained object recognition and zero-shot learning in remote sensing imagery / Gencer Sumbul in IEEE Transactions on geoscience and remote sensing, vol 56 n° 2 (February 2018)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 56 n° 2 (February 2018) . - pp 770 - 779 Titre : Fine-grained object recognition and zero-shot learning in remote sensing imagery Type de document : Article/Communication Auteurs : Gencer Sumbul, Auteur ; Ramazan Gokberk Cinbis, Auteur ; Selim Aksoy, Auteur Année de publication : 2018 Article en page(s) : pp 770 - 779 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] arbre urbain [Termes IGN] image numérique [Termes IGN] inférence [Termes IGN] reconnaissance de formes [Termes IGN] réseau neuronal convolutif Résumé : (Auteur) Fine-grained object recognition that aims to identify the type of an object among a large number of subcategories is an emerging application with the increasing resolution that exposes new details in image data. Traditional fully supervised algorithms fail to handle this problem where there is low between-class variance and high within-class variance for the classes of interest with small sample sizes. We study an even more extreme scenario named zero-shot learning (ZSL) in which no training example exists for some of the classes. ZSL aims to build a recognition model for new unseen categories by relating them to seen classes that were previously learned. We establish this relation by learning a compatibility function between image features extracted via a convolutional neural network and auxiliary information that describes the semantics of the classes of interest by using training samples from the seen classes. Then, we show how knowledge transfer can be performed for the unseen classes by maximizing this function during inference. We introduce a new data set that contains 40 different types of street trees in 1-ft spatial resolution aerial data, and evaluate the performance of this model with manually annotated attributes, a natural language model, and a scientific taxonomy as auxiliary information. The experiments show that the proposed model achieves 14.3% recognition accuracy for the classes with no training examples, which is significantly better than a random guess accuracy of 6.3% for 16 test classes, and three other ZSL algorithms. Numéro de notice : A2018-190 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2017.2754648 Date de publication en ligne : 18/10/2017 En ligne : https://doi.org/10.1109/TGRS.2017.2754648 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=89855 [article]