Further to previous work showing the superiority of the preprocessing algorithm developed by Gross and Brajovic, we propose improvements that remove the need for parameter selection. In extensive experimentation on the XM2VTS database, the Yale B database and the BANCA database, we show that our method of automatic parameter selection can produce better results than setting the parameter to a single value for the whole database. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Short, J., Kittler, J., & Messer, K. (2005). Photometric normalisation for face verification. In Lecture Notes in Computer Science (Vol. 3546, pp. 617–626). Springer Verlag. https://doi.org/10.1007/11527923_64
Mendeley helps you to discover research relevant for your work.