English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Paper

Self-supervised Multi-level Face Model Learning for Monocular Reconstruction at over 250 Hz

MPS-Authors
/persons/resource/persons206546

Tewari,  Ayush
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons136490

Zollhöfer,  Michael
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons127194

Garrido,  Pablo
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons214986

Bernard,  Florian
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons127713

Kim,  Hyeongwoo
Computer Graphics, MPI for Informatics, Max Planck Society;

/persons/resource/persons45610

Theobalt,  Christian
Computer Graphics, MPI for Informatics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)

arXiv:1712.02859.pdf
(Preprint), 4MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Tewari, A., Zollhöfer, M., Garrido, P., Bernard, F., Kim, H., Pérez, P., et al. (2017). Self-supervised Multi-level Face Model Learning for Monocular Reconstruction at over 250 Hz. Retrieved from http://arxiv.org/abs/1712.02859.


Cite as: http://hdl.handle.net/21.11116/0000-0000-615E-A
Abstract
The reconstruction of dense 3D models of face geometry and appearance from a single image is highly challenging and ill-posed. To constrain the problem, many approaches rely on strong priors, such as parametric face models learned from limited 3D scan data. However, prior models restrict generalization of the true diversity in facial geometry, skin reflectance and illumination. To alleviate this problem, we present the first approach that jointly learns 1) a regressor for face shape, expression, reflectance and illumination on the basis of 2) a concurrently learned parametric face model. Our multi-level face model combines the advantage of 3D Morphable Models for regularization with the out-of-space generalization of a learned corrective space. We train end-to-end on in-the-wild images without dense annotations by fusing a convolutional encoder with a differentiable expert-designed renderer and a self-supervised training loss, both defined at multiple detail levels. Our approach compares favorably to the state-of-the-art in terms of reconstruction quality, better generalizes to real world faces, and runs at over 250 Hz.