Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Texture characterization with semantic attributes: Database and algorithm

 
: Bormann, Richard; Esslinger, Dominik; Hundsdörfer, Daniel; Hägele, Martin; Vincze, Markus

:
Postprint urn:nbn:de:0011-n-4173583 (3.6 MByte PDF)
MD5 Fingerprint: c214d1513ac02eface75eed8d1a066fd
Created on: 11.10.2016


Verl, Alexander (Chairman, Tagungspräsident); Dragan, Mihai (Programmkomitee); Hägele, Martin (Programmkomitee) ; International Federation of Robotics; Deutsche Gesellschaft für Robotik -DGR-; Informationstechnische Gesellschaft -ITG-; Verband Deutscher Maschinen- und Anlagenbau e.V. -VDMA-, Fachverband Robotik und Automation, Frankfurt/Main; Fraunhofer-Institut für Produktionstechnik und Automatisierung -IPA-, Stuttgart:
47th International Symposium on Robotics 2016 : Robotics in the Era of Digitalization. June 21-22, 2016, Munich, Germany
Berlin: VDE-Verlag, 2016
ISBN: 978-3-8007-4231-8
pp.149-156
International Symposium on Robotics (ISR) <47, 2016, München>
English
Conference Paper, Electronic Publication
Fraunhofer IPA ()
semantisches Datenbankmodell; Algorithmus; Textur; Texturanalyse; Datenbankentwicklung; Roboter; texture description

Abstract
Categorization is an important capability of service robots for understanding their environment. While categorization may base on features of different modalities, such as 3D shape, size, or 2D interest points, this paper focuses on categorizing textures. In contrast to previous approaches that base texture classification on rather abstract numerical features, this work introduces a set of 17 properties that can easily be interpreted by a human establishing new options on human robot communication and learning. A new database was recorded to account for the service robotics context yielding almost 1500 images divided into 57 classes of textures and objects. The dataset is fully annotated with the 17 attributes which represent color and structural properties measured on a continuous scale. The evaluation on this database compares five methods for attribute learning, texture classification, and zero-shot learning with promising results.

: http://publica.fraunhofer.de/documents/N-417358.html