Hierarchical Rendering Techniques for Real-Time Approximate Illumination on Programmable Graphics Hardware

light pyramids

We introduce a GPU-based light hierarchy for real-time approximation of the illumination. We store information about virtual point lights such as position, direction and intensity into images. We then build a view-independent light hierarchy into quadtree image pyramids with a simple and rapid clustering algorithm. We approximate the illumination with small numbers of clusters (groups of lights) instead of large numbers of individual lights, using a new tree traversal algorithm on programmable graphics hardware.
    We also present an intelligent sub-sampling method to quickly generate images by exploiting spatial coherence. We make the image pyramids of the current screen buffers, and then we adaptively interpolate low-frequency regions with early-Z culling. We implemented our technique without occlusion, but we obtained visually-good results. We can control the quality–speed tradeoff through several scalar values. Entire steps run on programmable graphics hardware interactively without preprocessing.

Download:
- Paper

Selected Publication:
Hierarchical Rendering Techniques for Real-Time Approximate Illumination on Programmable Graphics Hardware Hyunwoo Ki. "Hierarchical Rendering Techniques for Real-Time Approximate Illumination on Programmable Graphics Hardware". Master's Thesis, Soongsil University, June 2007. (Note this thesis is an extended version of my Visual Computer journal paper. It introduces a hierarchical lighting as well as a hierarchical subsampling to accelerate final rendering.)

BibTeX:
@mastersthesis{KiH2007HRT,
author = {Hyunwoo Ki},
title = {Hierarchical rendering techniques for real-time approximate illumination on programmable graphics hardware},
school = {Soongsil University},
year = {2007},
month = {June}, }

Related Publication:
A GPU-Based Light Hierarchy for Real-Time Approximate Illumination Hyunwoo Ki, and Kyoungsu Oh. "A GPU-Based Light Hierarchy for Real-Time Approximate Illumination". The Visual Computer, Volume 24, Numbers 7-9, July, 2008, pp. 649-658.