View invariant head recognition by Hybrid PCA based reconstruction

Article type: Research Article

Authors: Wu, Qingquan | Ben-Arie, Jezekiel^{; *}

Affiliations: ECE Department, M/C 154, University of Illinois at Chicago, 851 S. Morgan St., Chicago IL 60607, USA

Correspondence: [*] Corresponding author. Tel.: +1 312 996 2648; Fax: +1 312 996 6465; E-mail: [email protected].

Abstract: We propose a novel method for 3D head reconstruction and view-invariant recognition from single 2D images. We employ a deterministic Shape From Shading (SFS) method with initial conditions estimated by Hybrid Principal Component Analysis (HPCA) and multi-level global optimization with error-dependent smoothness and integrability constraints. Our HPCA algorithm provides good initial estimates of 3D range mapping for the SFS optimization and yields much improved 3D head reconstruction. The paper also presents novel approaches to global optimization. It also describes a novel method in SFS handling of variable and unknown surface albedo, a problem with unsatisfactory solutions by prevalent SFS methods. In the experiments, we reconstruct 3D head range images from 2D single images in different views. The 3D reconstructions are then used to recognize stored model persons. This enables one to recognize faces in wide range of views. Empirical results show that our HPCA based SFS method provides 3D head reconstructions that notably improve the accuracy compared to other approaches. 3D reconstructions derived from images of 40 persons are tested against 80 3D head models and a recognition rate of over 90% is achieved. Such a capability was not demonstrated by any other method for view-invariant face recognition.

DOI: 10.3233/ICA-2008-15202

Journal: Integrated Computer-Aided Engineering, vol. 15, no. 2, pp. 97-108, 2008