Face recognition based on 2D images under illumination and pose variations

We propose a novel 2D image-based approach that can simultaneously handle illumination and pose variations to enhance face recognition rate. It is much simpler, requires much less computational effort than the methods based on 3D models, and provides a comparable or better recognition rate.

Research highlights
► The proposed method based on 2D images can handle illumination and pose variations.
► The feature space for each pose class are constructed for face recognition.
► The pose class of a probe image can be easily determined by using an edge image.
► The light direction category in the pose class is determined by using a binary image.
► It provides better recognition performance compared to other 3D model based methods.