This paper investigates the performance of eigenpaintings, SURF and a combination of both methods in a query-by- picture system dedicated to artwork images. For this purpose a database containing photographs of 17 artworks was created. These photos were taken by visitors of the exhibition {"}The VUB Art Collection - On Display and in Your Hand{"} using the camera of a mobile phone. The resulting images present several challenging aspects, including reflections, different light conditions, variations in perspective viewing conditions etc. Experiments show that the method based on eigenpaintings classifies 85% of the artworks correctly and the method based on SURF 84%. The combined classifier shows an accuracy of 88%.
Temmermans, F, Jansen, B, Deklerck, R, Schelkens, P & Cornelis, J 2011, The mobile museum guide: Artwork recognition with eigenpaintings and SURF. in WIAMIS 2011: 12th International Workshop on Image Analysis for Multimedia Interactive Services, Delft, The Netherlands, April 13-15, 2011. TU Delft; EWI; MM; PRB, Unknown, 13/04/11. <http://repository.tudelft.nl/view/conferencepapers/uuid%3A8f0dc78f-621e-42e8-8697-073d5b36111d/>
Temmermans, F., Jansen, B., Deklerck, R., Schelkens, P., & Cornelis, J. (2011). The mobile museum guide: Artwork recognition with eigenpaintings and SURF. In WIAMIS 2011: 12th International Workshop on Image Analysis for Multimedia Interactive Services, Delft, The Netherlands, April 13-15, 2011 TU Delft; EWI; MM; PRB. http://repository.tudelft.nl/view/conferencepapers/uuid%3A8f0dc78f-621e-42e8-8697-073d5b36111d/
@inproceedings{aad4f297cf1f4a1eac11ea337a70de7a,
title = "The mobile museum guide: Artwork recognition with eigenpaintings and SURF",
abstract = "This paper investigates the performance of eigenpaintings, SURF and a combination of both methods in a query-by- picture system dedicated to artwork images. For this purpose a database containing photographs of 17 artworks was created. These photos were taken by visitors of the exhibition {"}The VUB Art Collection - On Display and in Your Hand{"} using the camera of a mobile phone. The resulting images present several challenging aspects, including reflections, different light conditions, variations in perspective viewing conditions etc. Experiments show that the method based on eigenpaintings classifies 85% of the artworks correctly and the method based on SURF 84%. The combined classifier shows an accuracy of 88%.",
keywords = "mobile search, artwork recognition, surf, eigen paintings, image retrieval",
author = "Frederik Temmermans and Bart Jansen and Rudi Deklerck and Peter Schelkens and Jan Cornelis",
year = "2011",
month = apr,
day = "13",
language = "English",
isbn = "978-94-90818-00-5",
booktitle = "WIAMIS 2011: 12th International Workshop on Image Analysis for Multimedia Interactive Services, Delft, The Netherlands, April 13-15, 2011",
publisher = "TU Delft; EWI; MM; PRB",
note = "Unknown ; Conference date: 13-04-2011",
}