Technical Papers

Global Illumination

Wednesday, 24 July 9:00 AM - 10:30 AM
Session Chair: Holly Rushmeier, Yale University

Robust Adaptive Photon Tracing Using Photon-Path Visibility

A new photon-tracing method that takes the visibility of each photon path into account using an adaptive Markov chain Monte Carlo method.

Toshiya Hachisuka
Aarhus Universitet

Henrik Wann Jensen
University of California, San Diego

Adaptive Progressive Photon Mapping

A locally adaptive data‐driven error-minimization technique that optimally balances noise and bias in rendered images. The method derives asymptotic rates of convergence for progressive photon mapping.

Anton Kaplanyan
Karlsruher Institut für Technologie

Carsten Dachsbacher
Karlsruher Institut für Technologie

Gradient-Domain Metropolis Light Transport

This novel Metropolis rendering algorithm first computes image gradients along with a low-fidelity approximation of the image, and then reconstructs the final image by solving a Poisson equation. As an extension of path-space Metropolis light transport, the algorithm is well suited for difficult transport scenarios.

Jaakko Lehtinen
NVIDIA Research

Tero Karras
NVIDIA Research

Samuli Laine
NVIDIA Research

Miika Aittala
Aalto University, NVIDIA Research

Frédo Durand
Massachusetts Institute of Technology

Timo Aila
NVIDIA Research

Axis-Aligned Filtering for Interactive Physically Based Diffuse Indirect Lighting

This new algorithm for interactive rendering of physically based global illumination, based on a novel frequency analysis of indirect lighting, combines adaptive sampling by Monte Carlo path tracing with real-time reconstruction of the resulting noisy images. Analysis assumes diffuse indirect lighting, with general receiver BRDF.

Soham Uday Mehta
University of California, Berkeley

Brandon Wang
University of California, Berkeley

Ravi Ramamoorthi
University of California, Berkeley

Frédo Durand
Massachusetts Institute of Technology