In this work, we investigate how illuminant estimation techniques can be improved using indoor/outdoor classification. The illuminant estimation algorithms considered are derived from the framework recently proposed by Van de Weijer and Gevers. We have designed a strategy for the selection of the most appropriate algorithm on the basis of the classification results. We have tested the proposed strategy on a subset of the widely used Funt and Ciurea dataset. Experimental results clearly demonstrate that our strategy outperforms general purpose algorithms. © 2008 Springer-Verlag Berlin Heidelberg.

Schettini, R., Cusano, C., Ciocca, G., Bianco, S. (2008). Classification-based color constancy. In Visual Information Systems. Web-Based Visual Information Search and Management (pp.104-113). Springer [10.1007/978-3-540-85891-1_14].

Classification-based color constancy

SCHETTINI, RAIMONDO;CUSANO, CLAUDIO;CIOCCA, GIANLUIGI;BIANCO, SIMONE
2008

Abstract

In this work, we investigate how illuminant estimation techniques can be improved using indoor/outdoor classification. The illuminant estimation algorithms considered are derived from the framework recently proposed by Van de Weijer and Gevers. We have designed a strategy for the selection of the most appropriate algorithm on the basis of the classification results. We have tested the proposed strategy on a subset of the widely used Funt and Ciurea dataset. Experimental results clearly demonstrate that our strategy outperforms general purpose algorithms. © 2008 Springer-Verlag Berlin Heidelberg.
No
paper
classification, color constancy, algorithm selection
English
10th International Conference on Visual Information Systems
978-3-540-85890-4
Schettini, R., Cusano, C., Ciocca, G., Bianco, S. (2008). Classification-based color constancy. In Visual Information Systems. Web-Based Visual Information Search and Management (pp.104-113). Springer [10.1007/978-3-540-85891-1_14].
Schettini, R; Cusano, C; Ciocca, G; Bianco, S
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/12983
Citazioni
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 5
Social impact