A new face database and evaluation of face recognition techniques
Dimitrios S. Alexiadis, Vasileios Syrris, A. Papastergiou, A. T. Hatzigaidas, L. Mariuta · 2010
Abstract: This paper introduces ELA5, a new image database that is suitable for experimentation within the face recognition domain. Its design attempts to cover a variety of scenarios encompassing pose and illumination varia-tions, different facial expressions and occlusion. In addition, established computational techniques such as Princi-pal Component Analysis (PCA) and Multilinear PCA, combined with Fisher/Linear Discriminant Analysis (LDA) are evaluated and comparative results display their strengths and weaknesses in settings that simulate real-world conditions.