Non-Lambertian Reflectance Modeling and Shape Recovery for Faces using Anti-Symmetric Tensor Splines

by Kumar, R. | Barmpoutis, A. | Banerjee, A. | Vemuri, B. C.

IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33(3), March, 2011, pp. 533-567. https://doi.org/10.1109/TPAMI.2010.67

Category:
FacebookXEmailLinkedin

Description

Modeling illumination effects and pose variations of a face is of fundamental importance in the field of facial image analysis. Most of the conventional techniques that simultaneously address both of these problems work with the Lambertian assumption and thus, fall short of accurately capturing the complex intensity variation that the facial images exhibit or recovering their 3D shape in presence of specularities and cast shadows. In this paper we present a novel anti-symmetric tensor spline based framework for facial image analysis. We show that using this framework, facial apparent BRDF field can be accurately estimated while seamlessly accounting for cast shadows and specularities. Further, using local neighborhood information, the same framework can be exploited to recover the 3D shape of the face (to handle pose variation). We quantitatively validate the accuracy of the anti-symmetric tensor spline model using a more general continuous mixture of single lobed spherical functions. We demonstrate the effectiveness of our technique by presenting extensive experimental results for face relighting, 3D shape recovery and face recognition using the Extended Yale B and CMU PIE benchmark datasets.

Additional information

Author

Kumar, R., Barmpoutis, A., Banerjee, A., Vemuri, B. C.
Journal

IEEE Transactions on Pattern Analysis and Machine Intelligence
Volume

33
Number

3
Year

2011
Month

March
Pages

533-567
DOI

https://doi.org/10.1109/TPAMI.2010.67

Citation

Citation

Kumar, R., Barmpoutis, A., Banerjee, A. and Vemuri, B., 2011. Non-Lambertian Reflectance Modeling and Shape Recovery for Faces using Anti-Symmetric Tensor Splines. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33(3), pp. 533-567. https://doi.org/10.1109/TPAMI.2010.67

BibTex

@article{digitalWorlds:159,
doi = {https://doi.org/10.1109/TPAMI.2010.67},
author = {Kumar, R.  and  Barmpoutis, A.  and  Banerjee, A.  and  Vemuri, B. C.},
title = {Non-Lambertian Reflectance Modeling and Shape Recovery for Faces using Anti-Symmetric Tensor Splines},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
month = {March},
volume = {33},
number = {3},
year = {2011},
pages = {533-567}
}

See also…