Neural BRDF Representation and
Importance Sampling

• Alejandro Sztrajman
• Gilles Rainer
• Tobias Ritschel
• Tim Weyrich

Department of Computer Science, University College London, UK.
Computer Graphics Forum, 2021 (EGSR 2022).

Controlled capture of real-world material appearance yields tabulated sets of highly realistic reflectance data. In practice, however, its high memory footprint requires compressing into a representation that can be used efficiently in rendering while remaining faithful to the original. Previous works in appearance encoding often prioritised one of these requirements at the expense of the other, by either applying high-fidelity array compression strategies not suited for efficient queries during rendering, or by fitting a compact analytic model that lacks expressiveness. We present a compact neural network-based representation of BRDF data that combines high-accuracy reconstruction with efficient practical rendering via built-in interpolation of reflectance. We encode BRDFs as lightweight networks, and propose a training scheme with adaptive angular sampling, critical for the accurate reconstruction of specular highlights. Additionally, we propose a novel approach to make our representation amenable to importance sampling: rather than inverting the trained networks, we learn an embedding that can be mapped to parameters of an analytic BRDF for which importance sampling is known. We evaluate encoding results on isotropic and anisotropic BRDFs from multiple real-world datasets, and importance sampling performance for isotropic BRDFs mapped to two different analytic models.

• Paper
  PDF 16 MB
• Paper (lowres)
  PDF 3.1 MB
• Supplemental MERL
  PDF 55 MB

• Presentation Slides
  PDF 5.9 MB
• Code
  

WebGL Demo

Code and Pre-trained Models

Code is available in the github repository. We provide pre-trained NBRDFs for measured materials from multiple databases:

• Python code to train NBRDF networks.
• Mitsuba plugin (C++) to render using pre-trained NBRDFs.
• All materials from the MERL database.
• Isotropic materials from the RGL database.
• Materials from Nielsen et al. (2015).
• NBRDF training script (alternative implementation in PyTorch by Michael Fischer).

xxxxxxxxxx
@article{sztrajman2021nbrdf,
    author={Alejandro Sztrajman and Gilles Rainer and Tobias Ritschel and Tim Weyrich},
    title = {Neural BRDF Representation and Importance Sampling},
    journal = {Computer Graphics Forum},
    year = {2021},
    doi = {https://doi.org/10.1111/cgf.14335},
    url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/cgf.14335}
}