Cascade classifier for face detection

Detecting faces in images is a key step in numerous computer vision applications, such as face recognition or facial expression analysis. Automatic face detection is a difficult binary classification problem because of the large face intra-class variability which is due to the important influence of the environmental conditions on the face appearance. We propose a cascade face detection method based on histograms of oriented gradients (HOG), using different kinds of features and classifier to exclude non-face step by step. The candidate feature set was constructed by HOG feature of different grain size; at different stages the support vector machine (SVM) was used as the weak classifier with different parameters. Experimental results showed a better performance compared to the state-of-the-art on Carnegie Mellon University/ Massachusetts Institute of Technology (CMU/MIT) datasets.