Multimodal Face Recognition System Using Spectral Transformation of 2D Texture Feature and Statistical Processing of Face Range Images
S. Naveen, R. S. Moni · Procedia Computer Science · 2015
3D Face recognition has been an area of interest for the past few decades in pattern recognition. This paper focuses on problems of person identification using 3D Face data. Here unregistered Face data, i.e. both texture and depth is fed to classifier in spectral representations of data. 2D Discrete Fourier Transform (DFT) is used for spectral representation. Fusion of scores improves the recognition accuracy significantly since use of depth information alone in spectral representation was not sufficient to increase accuracy. Statistical method seems to degrade performance of system when applied to texture data and was effective for depth data.