GratNet: A Photorealistic Neural Shader for Diffractive Surfaces

TL;DR

GratNet introduces a neural shader that efficiently renders diffractive surfaces with high fidelity, significantly reducing data requirements and outperforming traditional wave optics models in quality and speed.

Contribution

This paper presents a novel MLP-based neural rendering method for diffractive surfaces that minimizes data dependency and enhances efficiency compared to existing wave-optical approaches.

Findings

Achieves high-quality surface reconstructions with PSNR, SSIM, and FLIP metrics.

Reduces raw dataset memory footprint by two orders of magnitude.

Produces visually similar renderings to state-of-the-art methods with better performance.

Abstract

Structural coloration is commonly modeled using wave optics for reliable and photorealistic rendering of natural, quasi-periodic and complex nanostructures. Such models often rely on dense, preliminary or preprocessed data to accurately capture the nuanced variations in diffractive surface reflectances. This heavy data dependency warrants implicit neural representation which has not been addressed comprehensively in the current literature. In this paper, we present a multi-layer perceptron (MLP) based method for data-driven rendering of diffractive surfaces with high accuracy and efficiency. We primarily approach this problem from a data compression perspective to devise a nuanced training and modeling method which is attuned to the domain and range characteristics of diffractive reflectance datasets. Importantly, our approach avoids over-fitting and has robust resampling behavior. Using Peak-Signal-to-Noise (PSNR), Structural Similarity Index Measure (SSIM) and a flipping difference evaluator (FLIP) as evaluation metrics, we demonstrate the high-quality reconstruction of the ground-truth. In comparison to a recent state-of-the-art offline, wave-optical, forward modeling approach, our method reproduces subjectively similar results with significant performance gains. We reduce the memory footprint of the raw datasets by two orders of magnitude in general. Lastly, we depict the working of our method with actual surface renderings.