Copula eigenfaces with attributes: Semiparametric principal component analysis for a combined color, shape and attribute model
Egger B, Kaufmann D, Schonborn S, Roth V, Vetter T (2017)
Publication Type: Conference contribution
Publication year: 2017
Journal
Publisher: Springer Verlag
Book Volume: 693
Pages Range: 95-112
Conference Proceedings Title: Communications in Computer and Information Science
Event location: Rome
ISBN: 9783319648699
DOI: 10.1007/978-3-319-64870-5_5
Abstract
Principal component analysis is a ubiquitous method in parametric appearance modeling for describing dependency and variance in datasets. The method requires the observed data to be Gaussian-distributed. We show that this requirement is not fulfilled in the context of analysis and synthesis of facial appearance. The model mismatch leads to unnatural artifacts which are severe to human perception. As a remedy, we use a semiparametric Gaussian copula model, where dependency and variance are modeled separately. This model enables us to use arbitrary Gaussian and non-Gaussian marginal distributions. Moreover, facial color, shape and continuous or categorical attributes can be analyzed in an unified way. Accounting for the joint dependency between all modalities leads to a more specific face model. In practice, the proposed model can enhance performance of principal component analysis in existing pipelines: The steps for analysis and synthesis can be implemented as convenient pre- and post-processing steps.
Authors with CRIS profile
Involved external institutions
Switzerland (CH)How to cite
APA:
Egger, B., Kaufmann, D., Schonborn, S., Roth, V., & Vetter, T. (2017). Copula eigenfaces with attributes: Semiparametric principal component analysis for a combined color, shape and attribute model. In Alexandru Telea, Paul Richard, Lars Linsen, Jose Braz, Sebastiano Battiato, Nadia Magnenat-Thalmann, Francisco Imai (Eds.), Communications in Computer and Information Science (pp. 95-112). Rome, IT: Springer Verlag.
MLA:
Egger, Bernhard, et al. "Copula eigenfaces with attributes: Semiparametric principal component analysis for a combined color, shape and attribute model." Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2016, Rome Ed. Alexandru Telea, Paul Richard, Lars Linsen, Jose Braz, Sebastiano Battiato, Nadia Magnenat-Thalmann, Francisco Imai, Springer Verlag, 2017. 95-112.
BibTeX: Download