Spherical Gaussian Light‐field Textures for Fast Precomputed Global Illumination
| Autor: | Dan Dolonius, Ulf Assarsson, Erik Sintorn, Roc Ramon Currius |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Optimization problem
Global illumination Computer science business.industry Gaussian 020207 software engineering 02 engineering and technology Computer Graphics and Computer-Aided Design Convolutional neural network symbols.namesake Amplitude 0202 electrical engineering electronic engineering information engineering symbols 020201 artificial intelligence & image processing Ray tracing (graphics) Computer vision Artificial intelligence business Light field Interpolation |
| Zdroj: | Computer Graphics Forum. 39:133-146 |
| ISSN: | 1467-8659 0167-7055 |
| DOI: | 10.1111/cgf.13918 |
| Popis: | We describe a method to use Spherical Gaussians with free directions and arbitrary sharpness and amplitude to approximate the precomputed local light field for any point on a surface in a scene. This allows for a high‐quality reconstruction of these light fields in a manner that can be used to render the surfaces with precomputed global illumination in real‐time with very low cost both in memory and performance. We also extend this concept to represent the illumination‐weighted environment visibility , allowing for high‐quality reflections of the distant environment with both surface‐material properties and visibility taken into account. We treat obtaining the Spherical Gaussians as an optimization problem for which we train a Convolutional Neural Network to produce appropriate values for each of the Spherical Gaussians' parameters. We define this CNN in such a way that the produced parameters can be interpolated between adjacent local light fields while keeping the illumination in the intermediate points coherent. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
| Nepřihlášeným uživatelům se plný text nezobrazuje | K zobrazení výsledku je třeba se přihlásit. |