Figure 4From: Handling missing weak classifiers in boosted cascade: application to multiview and occluded face detection Computation of P knn ( y =1|x) . Two weak classifiers are available: h a 1 and h a 2 . Applying these weak classifiers on a training database gives the set of points ℋ a (the red circles are positive points and the blue squares are negative points). Given an unknown sample x, h a 1 (x) and h a 2 (x) are computed and the k nearest neighbors of ( h a 1 (x), h a 2 (x)) are searched in ℋ a . In the figure, k=3, and the nearest neighbors are two positive points and one negative point which leads to P knn(y=1|x)=2/3.Back to article page