20 March 2013 Appearance-based color face recognition with 3D model
Author Affiliations +
Proceedings Volume 8768, International Conference on Graphic and Image Processing (ICGIP 2012); 87680J (2013) https://doi.org/10.1117/12.2003304
Event: 2012 International Conference on Graphic and Image Processing, 2012, Singapore, Singapore
Appearance-based face recognition approaches explore color cues of face images, i.e. grey or color information for recognition task. They first encode color face images, and then extract facial features for classification. Similar to conventional singular value decomposition, hypercomplex matrix also exists singular value decomposition on hypercomplex field. In this paper, a novel color face recognition approach based on hypercomplex singular value decomposition is proposed. The approach employs hypercomplex to encode color face information of different channels simultaneously. Hypercomplex singular value decomposition is utilized then to compute the basis vectors of the color face subspace. To improve learning efficiency of the algorithm, 3D active deformable model is exploited to generate virtual face images. Color face samples are projected onto the subspace and projection coefficients are utilized as facial features. Experimental results on CMU PIE face database verify the effectiveness of the proposed approach.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chengzhang Wang, Chengzhang Wang, Xiaoming Bai, Xiaoming Bai, } "Appearance-based color face recognition with 3D model", Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 87680J (20 March 2013); doi: 10.1117/12.2003304; https://doi.org/10.1117/12.2003304


Back to Top