Combining Language Sources and Robust Semantic Relatedness for Attribute-Based Knowledge Transfer

  • Marcus Rohrbach
  • Michael Stark
  • György Szarvas
  • Bernt Schiele
Conference paper

DOI: 10.1007/978-3-642-35749-7_2

Volume 6553 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Rohrbach M., Stark M., Szarvas G., Schiele B. (2012) Combining Language Sources and Robust Semantic Relatedness for Attribute-Based Knowledge Transfer. In: Kutulakos K.N. (eds) Trends and Topics in Computer Vision. ECCV 2010. Lecture Notes in Computer Science, vol 6553. Springer, Berlin, Heidelberg

Abstract

Knowledge transfer between object classes has been identified as an important tool for scalable recognition. However, determining which knowledge to transfer where remains a key challenge. While most approaches employ varying levels of human supervision, we follow the idea of mining linguistic knowledge bases to automatically infer transferable knowledge. In contrast to previous work, we explicitly aim to design robust semantic relatedness measures and to combine different language sources for attribute-based knowledge transfer. On the challenging Animals with Attributes (AwA) data set, we report largely improved attribute-based zero-shot object class recognition performance that matches the performance of human supervision.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Marcus Rohrbach
    • 1
    • 2
  • Michael Stark
    • 1
    • 2
  • György Szarvas
    • 1
  • Bernt Schiele
    • 1
    • 2
  1. 1.Department of Computer ScienceTU DarmstadtGermany
  2. 2.Max Planck Institute for InformaticsSaarbrückenGermany