Multiple And Single Haar Classifier For Face Recognition

Abstract

Haar-Classifier is a well-known tool for face detection in an image. Open-CV library already provide a class for Haar-Classifier implementation. However, one significant problem in face detection is tilt-face-pose recognition. In this paper, we will discuss about two different solutions to overcome the problem. The solutions are written in Python which has a very simple syntax. The first solution is using multiple classifier. Each of them is trained to recognize face with different rotation-degrees. The second solution is using single classifier to recognize many face image which have been rotated by different angles.