Robust Approaches for Multi-Label Face Classification

Ahmed Abdulateef Mohammed, Atul Sajjanhar

Abstract: Texture analysis is widely used to extract facial attributes which are used for face classification. In this paper, novel approaches for texture analysis are proposed to improve the performance of face classification. The proposed methods are derived by applying Local Binary Pattern based approaches on polar raster sampled face images. The proposed methods are evaluated by performing classification of face images using multi-label classification models. Experimental results show that the proposed methods outperform Local Binary Pattern method and its variants.