Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/30363
Appears in Collections:Computing Science and Mathematics Conference Papers and Proceedings
Author(s): Penatti, Otavio A B
Nogueira, Keiller
dos Santos, Jefersson A
Title: Do deep features generalize from everyday objects to remote sensing and aerial scenes domains?
Citation: Penatti OAB, Nogueira K & dos Santos JA (2015) Do deep features generalize from everyday objects to remote sensing and aerial scenes domains?. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2015, Boston, MA, USA, 07.06.2015-12.06.2015. Piscataway, NJ, USA: IEEE. https://doi.org/10.1109/cvprw.2015.7301382
Issue Date: Jun-2015
Conference Name: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2015
Conference Dates: 2015-06-07 - 2015-06-12
Conference Location: Boston, MA, USA
Abstract: In this paper, we evaluate the generalization power of deep features (ConvNets) in two new scenarios: aerial and remote sensing image classification. We evaluate experimentally ConvNets trained for recognizing everyday objects for the classification of aerial and remote sensing images. ConvNets obtained the best results for aerial images, while for remote sensing, they performed well but were outperformed by low-level color descriptors, such as BIC. We also present a correlation analysis, showing the potential for combining/fusing different ConvNets with other descriptors or even for combining multiple ConvNets. A preliminary set of experiments fusing ConvNets obtains state-of-the-art results for the well-known UCMerced dataset.
Status: VoR - Version of Record
Rights: This CVPR2015 workshop paper is the Open Access version, provided by the Computer Vision Foundation. The authoritative version of this paper is available in IEEE Xplore: https://doi.org/10.1109/CVPRW.2015.7301382

Files in This Item:
File Description SizeFormat 
10.1.1.883.8108.pdfFulltext - Published Version2.91 MBAdobe PDFView/Open



This item is protected by original copyright



Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.