Bidirectional Texture Function

In this project, we investigate the visual appearance of real-world surfaces and the dependence of appearance on the illumination and imaging geometry. We present a new texture representation called the BTF (Bidirectional Texture Function) which captures the variation in texture with illumination and viewing direction. We have developed a BTF database with image textures from over 60 different samples, each observed with over 200 different combinations of viewing and illumination directions. A related quantity to the BTF is the familiar BRDF (Bidirectional Reflectance Distribution Function). The measurement methods involved in the BTF database are conducive to a simultaneous measurement of the BRDF. Hence, we also present a BRDF database with reflectance measurements for over 60 different samples, each observed with over 200 different combinations of viewing and illumination directions. Both of these databases (BTF and BRDF) are publicly available and have important implications for computer vision and computer graphics. We have also developed histogram and correlation models for representing BTFs and used these models to develop texture recognition algorithms that are invariant to illumination and viewing directions.

Publications

"Multiresolution Histograms and their use for Texture Classification,"
E. Hadjidemetriou, M.D. Grossberg and S.K. Nayar,
3rd International Workshop on Texture Analysis and Synthesis (Texture 2003),
Oct, 2003.
[PDF] [bib] [©]

"Recognition of Dynamic Textures using Impulse Responses of State Variables,"
K. Fujita and S.K. Nayar,
3rd International Workshop on Texture Analysis and Synthesis (Texture 2003),
Oct, 2003.
[PDF] [bib] [©]

"Correlation Model for 3D Texture,"
K.J. Dana and S.K. Nayar,
IEEE International Conference on Computer Vision (ICCV),
Vol.2, pp.1061-1066, Sep, 1999.
[PDF] [bib] [©]

"Bidirectional Reflectance Distribution Function of Thoroughly Pitted Surfaces,"
J.J. Koenderink, A.J. Van Doorn, K.J. Dana and S.K. Nayar,
International Journal on Computer Vision,
Vol.31, No.2/3, pp.129-144, Apr, 1999.
[PDF] [bib] [©]

"Texture Histograms as a Function of Irradiation and Viewing Direction,"
B. Van-Ginneken, J.J. Koenderink and K.J. Dana,
International Journal on Computer Vision,
Vol.31, No.2/3, pp.169-184, Apr, 1999.
[PDF] [bib] [©]

"Reflectance and Texture of Real World Surfaces,"
K.J. Dana, B. Van-Ginneken, S.K. Nayar, J.J. Koenderink,
ACM Transactions on Graphics (TOG),
Vol.18, No.1, pp.1-34, Jan, 1999.
[PDF] [bib] [©]

"Histogram Model for 3D Textures,"
K.J. Dana and S.K. Nayar,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
pp.618-624, Jun, 1998.
[PDF] [bib] [©]

"Reflectance and Texture of Real World Surfaces,"
K.J. Dana, B.V. Ginneken, S.K. Nayar and J.J. Koenderink,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
pp.151-157, Jun, 1997.
[PDF] [bib] [©]

"Reflectance and Texture of Real World Surfaces,"
K.J. Dana, B. van Ginneken, S.K. Nayar and J.J. Koenderink,
Technical Report, Department of Computer Science, Columbia University CUCS-048-96,
1996.
[PDF] [bib] [©]

"Generalization of the Lambertian Model and Implications for Machine Vision,"
S.K. Nayar and M. Oren,
International Journal on Computer Vision,
Vol.14, No.3, pp.227-251, Apr, 1995.
[PDF] [bib] [©]

"Visual Appearance of Matte Surfaces,"
S.K. Nayar and M. Oren,
Science,
Vol.267, pp.1153-1156, Feb, 1995.
[PDF] [bib] [©]

Image

  Cover of ACM TOG:

This picture shows a cylinder rendered using texture mapping which assumes the texture to be two-dimensional (left column) and using an algorithm that assumes the texture to be three-dimensional (right column). These examples show why the bidirectional texture function is key to achieving photorealism in computer graphics.

     

Database

CURET: Reflectance and Texture Database

Oren-Nayar Reflectance Model

Appearance Matching

Histograms: Properties and Applications