EON: A practical energy-preserving rough diffuse BRDF

Abstract: We introduce the "Energy-preserving Oren--Nayar" (EON) model for reflection from rough surfaces. Unlike the popular qualitative Oren--Nayar model (QON) and its variants, our model is energy-preserving via analytical energy compensation. We include self-contained GLSL source code for efficient evaluation of the new model and importance sampling based on a novel technique we term "Clipped Linearly Transformed Cosine" (CLTC) sampling.

• The paper presents the EON model as an energy-preserving BRDF that outperforms traditional Oren–Nayar models by effectively addressing energy loss in high-order scattering.
• It employs a novel Kelemen-like lobe and CLTC importance sampling, delivering up to a 7x speedup and over 100x variance reduction at grazing angles.
• The model’s efficient, straightforward implementation, complete with GLSL code, offers immediate benefits for realistic rendering and further research in scattering models.

An Academic Analysis of the EON Model for Rough Diffuse BRDF

The paper presents the Energy-preserving Oren–Nayar (EON) model, a novel approach to modeling reflection from rough surfaces, building on the foundational concepts introduced by Oren and Nayar. The EON model aims to preserve energy more effectively than its predecessors while maintaining practical applicability for rendering and graphics applications.

Overview of Existing Models

The traditional Oren–Nayar models, including the fullON and QON versions, provide significant insights into surface reflectance but suffer from energy loss issues and complexity in computation. These models are not energy preserving, leading to discrepancies in rendering scenarios, particularly at grazing angles and high roughness values.

Fujii's modifications attempted to approximate the full model while addressing some of these energy issues, but limitations remained, particularly in efficiently handling high-order scattering events.

Introduction of the EON Model

The EON model is explicitly designed to be both energy-preserving and straightforward to implement. This is achieved by integrating a compensation term, inspired by previous energy compensation techniques used in microfacet models for conductors. The model includes a Kelemen-like lobe to account for missing energy due to higher order scattering events.

The EON model is highly efficient and rectifies artifacts found in other models, such as the dark ring in QON. It uses a novel approach to adjust the scattering coefficients based on roughness, allowing it to maintain visual fidelity across varying conditions.

Numerical Results and Implementation

The EON model's practical performance is highlighted by its approximate sevenfold efficiency improvement over the fullON model. The paper provides detailed GLSL code for implementation, addressing concerns of computational efficiency while preserving the model's integrity.

Importance Sampling and Variance Reduction

A significant contribution of the paper is the development of a novel importance sampling strategy for the EON model. The introduction of Clipped Linearly Transformed Cosine (CLTC) sampling allows for more accurate sampling distributions, reducing variance by a factor exceeding 100 at grazing angles compared to cosine-weighted sampling. This importance sampling method is critical for practical rendering scenarios where efficiency and noise reduction are paramount.

Theoretical and Practical Implications

The EON model's introduction has substantial implications for both theoretical and applied graphics research. Theoretically, it provides a more accurate representation of rough surface interactions by addressing energy conservation issues inherent in previous models. Practically, its efficient implementation makes it suitable for high-performance graphics applications, evidenced by its adoption in the OpenPBR specification.

Future Directions

The paper suggests further exploration could include aligning the EON model more closely with microfacet theories, enhancing its physical accuracy. Additionally, there is potential to adapt the CLTC sampling scheme for alternative BRDF models, such as the VMF diffuse model, to further enhance rendering realism.

In summary, the EON model stands as a well-rounded enhancement in the domain of BRDF modeling, addressing critical challenges in energy conservation, computational efficiency, and visual realism. The discussions in the paper open avenues for further research into refined scattering models and improved sampling techniques, promising advancements in realistic rendering and simulation.