2007/012 - Safe Red-Eye corrrection plug-in using adaptive methods
Luca Marchesotti, Gabriela Csurka, Marco Bressan
Computational Color Imaging Workshop (CCIW 07), Modena, Italy, September 14, 2007.
We introduce a new method for the correction of red-eyes in photography. An important issue with red-eye correction is that it might result in image degradation. This can be due to the detection of false positives or, even in the case of correct detection to an inadequate correction technique. An evaluation of correction strategies is carried out taking into account the risk of image degradation and the expected perceptual quality improvement. Different correction strategies are presented and integrated in a plug-in that builds up on performances of any red-eye detector which satisfies some minimum requirements.