A Data-Driven Paradigm for Precomputed Radiance Transfer

TL;DR

This paper introduces a data-driven approach to Precomputed Radiance Transfer (PRT) that simplifies the process and enables real-time rendering of indirect illumination using machine learning principles and standard linear algebra tools.

Contribution

It presents a novel data-driven paradigm for PRT that reduces complexity and facilitates real-time indirect illumination rendering with a simple baseline method.

Findings

Real-time rendering of indirect illumination achieved.

Uses standard tools like SVD for basis and transfer function extraction.

Simplifies traditional PRT construction process.

Abstract

In this work, we explore a change of paradigm to build Precomputed Radiance Transfer (PRT) methods in a data-driven way. This paradigm shift allows us to alleviate the difficulties of building traditional PRT methods such as defining a reconstruction basis, coding a dedicated path tracer to compute a transfer function, etc. Our objective is to pave the way for Machine Learned methods by providing a simple baseline algorithm. More specifically, we demonstrate real-time rendering of indirect illumination in hair and surfaces from a few measurements of direct lighting. We build our baseline from pairs of direct and indirect illumination renderings using only standard tools such as Singular Value Decomposition (SVD) to extract both the reconstruction basis and transfer function.