Real-time face detection using Gentle AdaBoost and nesting cascade structure

In this paper, a fast face detector based on Gentle AdaBoost algorithm and nesting cascade structure is proposed. In order to solve the problem that weak classifiers with same features often appear in different node classifiers of a classical cascade classifier to slow down the face detection speed due to its independently training each node, the nesting cascade structure is introduced to connect the node classifiers, and the features used in the previous node classifier are removed from a Haar-like feature set during the training process. Gentle AdaBoost is used to train node classifiers on the Haar-like feature set to improve the generalization ability of the node classifier, which will further reduce the number of weak classifier in nesting cascade structure. Experimental results have proved that the proposed algorithm can significantly reduce the number of weak classifiers, increase the detection speed, and slightly raise the detection accuracy as well. On the CIF(352×288) video, the average detection speed of the proposed face detector can achieve 125fps, which is superior to state-of-the arts of face detection and completely satisfies the demand of real-time face detection.