Compact Representation of Reflectance Fields using Clustered Sparse Residual Factorization
ICT Technical Report, ICT-TR-02-2009, June 2009
| Hideshi Yamada2 Pieter Peers1 Paul Debevec1 |
| University of Southern California, Institute for Creative Technologies1 Sony Corporation2 |
Relit results of a reflectance field factorized with 2 terms and varying residual terms (left to right: 0, 2.77, and 4.74 procent). Each inset shows the difference with a reference relit solution.
Abstract:
We present a novel compression method for fixed viewpoint reflectance fields, captured for example by a Light Stage. Our compressed representation consists of a global approximation that exploits the similarities between the reflectance functions of different pixels, and a local approximation that encodes the per-pixel residual with the global approximation. Key to our method is a clustered sparse residual factorization. This sparse residual factorization ensures that the per-pixel residual matrix is as sparse as possible, enabling a compact local approximation. Finally, we demonstrate that the presented compact representation is well suited for high-quality real-time rendering.