Finding Human Faces in a Picture

We present a multiscale ruled-based algorithm for detecting human faces in still monochromatic images with complex background. To reduce search time, three coarse-to-fine levels are used. At each level, face candidates are picked or rejected according to a small set of rules which are hand-crafted from training samples. Preliminary results are encouraging and indicate that our algorithm compares favorably with existing methods.

[1]  Venu Govindaraju,et al.  A Computational Model for Face Location Based on Cognitive Principles , 1992, AAAI.

[2]  Alex Pentland,et al.  Face recognition using eigenfaces , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[3]  M. Kosugi.  Robust identification of human face using mosaic pattern and BPN , 1992, Neural Networks for Signal Processing II Proceedings of the 1992 IEEE Workshop.