Detection of Glasses in Facial Images

In this paper we consider the automatic detection of the presence of glasses. Using the eye locations detected in facial images, a region below the eyes is investigated where glasses are potentially expected. We introduce three measures for the likelihood of glasses. Experimental results have demonstrated a good separation of the two classes of facial images with and without glasses, based on these measures. Information about the presence of glasses will be useful to support automatic annotation of large image databases and indexing for fast face recognition.

[1]  Alex Pentland,et al.  View-based and modular eigenspaces for face recognition , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Niels da Vitoria Lobo,et al.  Age classification from facial images , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[3]  Ravi Kothari,et al.  Detection of eye locations in unconstrained visual images , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[4]  A. J. O'toole,et al.  Classifying faces by face and sex using an autoassociative memory trained for recognition , 1991 .

[5]  Terrence J. Sejnowski,et al.  Sex Recognition from Faces Using Neural Networks , 1995 .

[6]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .