Dealing with multi-view faces is important for many face-related applications. Statistics show that approximately 75% of the faces in home photos are non-frontal. Multi-view face detection has been a challenging problem. The challenge is firstly due to large amount of variation and complexity brought about by the changes facial appearance, lighting and expression. Changes in facial view (pose) further complicate the situation because the distribution of multi-view faces in a feature space is more dispersed and more complicated than that of frontal faces. A new boosting algorithm, called FloatBoost, is proposed to overcome the monotonicity problem of the sequential AdaBoost learning. AdaBoost is a sequential forward search procedure using the greedy selection strategy. The premise offered by the sequential procedure can be broken-down when the monotonicity assumption, i.e. that when adding a new feature to the current set, the value of the performance criterion does not decrease, is violated. FloatBoost incorporates the idea of Floating Search into AdaBoost to solve the non-monotonicity problem encountered in the We then present a system which learns to detect multi-view faces using FloatBoost. The system uses a coarse-to-fine, simple-to-complex architecture called detector-pyramid. FloatBoost learns the component detectors in the pyramid and yields similar or higher classification accuracy that AdaBoost withy a smaller number of weak classifiers. This work leads to the first real-time multi-view face detection system in the world. It runs at 200 ms per image of size 320*240 pixels on a Pentimu-Ⅲ CPU of 700 MHz.
修改评论