Pose invariant face recognition based on hybrid-global linear regression

Abstract

The paper presents a simple but efficient novel H-eigenface (Hybrid-eigenface) method for pose invariant face recognition ranging from frontal to profile view. H-eigenfaces are entirely new basis for face image representation under different poses and are used for virtual frontal view synthesis. The proposed method is based on the fact that face samples of same person under different poses are similar in terms of the combination pattern of facial features. H-eigenfaces exploit this fact and thus two H-eigenfaces under different poses capture same features of the face. Thereby providing a compact view-based subspace, which can be further used to generate virtual frontal view from inputted non-frontal face image using least square projection technique. The use of proposed methodology on FERET and ORL face database shows an impressive improvement in recognition accuracy and a distinct reduction in online computation when compared to global linear regression method.

Appendix: universal image quality index

The quality index Q of a test image $\left( {{\mathbf{t}} = \left\{ {t_{i} |i = 1,2, \ldots, N} \right\}} \right)$ with respect to the reference image $\left( {r = \left\{ {r_{i} |i = 1,2, \ldots, N} \right\}} \right)$ is defined as-

$$ Q = {\frac{{4\sigma_{tr} \tilde{t}\tilde{r}}}{{(\sigma_{t}^{2} + \sigma_{r}^{2} )(\tilde{t}^{2} + \tilde{r}^{2} )}}} $$

where,

$$ \begin{gathered} \begin{array}{*{20}c} {\tilde{t} = \frac{1}{N}\sum\limits_{i = 1}^{N} {t_{i} } ,} & {\tilde{r} = \frac{1}{N}\sum\limits_{i = 1}^{N} {r_{i} } ,} & {\sigma_{t}^{2} = {\frac{1}{N - 1}}\sum\limits_{i = 1}^{N} {(t_{i} - \tilde{t})^{2} } ,} \\ \end{array} \hfill \\ \begin{array}{*{20}c} {\sigma_{r}^{2} = {\frac{1}{N - 1}}\sum\limits_{i = 1}^{N} {(r_{i} - \tilde{r})^{2} ,} } & {\sigma_{rt}^{2} = {\frac{1}{N - 1}}\sum\limits_{i = 1}^{N} {(r_{i} - \tilde{r})(t_{i} - \tilde{t})} } \\ \end{array} . \hfill \\ \end{gathered} $$