Post on 21-Dec-2015
transcript
4EyesFace-Realtime face detection, tracking, alignment and recognition
Changbo Hu, Rogerio Feris and Matthew Turk
Introduction
Why this is a difficult problem? Facial Expressions, Illumination Changes, Pose,
etc. Object
Develop a fully automatic system, suitable for real-time applications to locate and track human faces, then to align and recognize the face.
Evaluate it on a large dataset.
Face Detection
[Viola and Jones, 2001]
Simple features, which can be computed very fast.
A variant of Adaboost is used both to select the features and to train the classifier.
Classifiers are combined in a “cascade” which allows background regions of the image to be quickly discarded.
Face Alignment
Active Appearance Model (AAM)
Statistical Shape Model (PCA)
Statistical Texture Model (PCA)
Face alignment
Problem: Partial Occlusion
Active Wavelet Networks (AWN) (on BMVC’03) Main idea: Replace AAM texture model by a
wavelet network
Face Alignment
Similar performance to AAM in images under normal conditions.
More robust against partial occlusions.
Face Alignment
Using 9 wavelets, the system requires only 3 ms per iteration. In general, at most 10 iterations are sufficiently for good convergence (PIV 1.6Ghz).
Face recognition
Large dataset evaluation FERET DataSet 1196 different individuals1196 different individuals With ground truth of eye cornersWith ground truth of eye corners