Kernel TV-Based Quotient Image Employing Gabor Analysis and Its Application to Face Recognition

Abstract

SUMMARY In order to overcome the drawback of TVQI and to utilize the property of dimensionality increasing techniques, a novel model for Kernel TV-based Quotient Image employing Gabor analysis is proposed and applied to face recognition with only one sample per subject. To deal with illumination outliers, an enhanced TV-based quotient image (ETVQI) model is first adopted. Then for preprocessed images by ETVQI, a bank of Gabor filters is built to extract features at specified scales and orientations. Lastly, KPCA is introduced to extract final high-order and nonlinear features of extracted Gabor features. According to experiments on the CAS-PEAL face database, our model could outperform Gabor-based KPCA, TVQI and Gabor-based TVQI when they face most outliers (illumination, expression, masking etc.).

Illumination Normalization using Quotient Image-based Techniques

“…Our aim is to do this by synthesizing an illumination normalized image using Quotient Image-based techniques (Shashua et al, 2001, Wang et al, 2004, Chen et al, 2005, Nishiyama et al, 2006, Zhang et al, 2007, and An et al, 2008. These techniques extract an illumination invariant representation of a face from a raw facial image.…”

“…The SQI however is neither synthesized at the boundary between a diffuse reflection region and a shadow region, nor at the boundary between a diffuse reflection region and a specular reflection region. This problem is not addressed in the Total Variation Quotient Image (Chen et al, 2005, An et al, 2008 and the Morphological Quotient Image (Zhang et al, 2007). Determining the reflectance type of an appearance from a single image is an ill-posed problem.…”

Independent Gabor Analysis of Multiscale Total Variation-Based Quotient Image

“…However, most important large-scale features, which contain information of cheekbone, muscle and profile, etc., are ignored by TVQI. These large-scale features would play an important part in the whole recognition procedure [8], [9]. Table I In this letter, a new model for independent Gabor analysis of multiscale TV-based quotient image is proposed and applied to face recognition with only one sample per subject.…”

An Active Illumination and Appearance (AIA) Model for Face Alignment

“…Wang et al [9] developed reflectance estimation methods by using the idea of the ratio of the original image and its smooth version. In the same research direction Zhang [10] and An [11] proposed new methods to extract an illumination invariant representation of a face images from a raw facial images.…”