2023•Proceedings of the ACM on Computer Graphics and Interactive TechniquesRequires access
Importance-Based Ray Strategies for Dynamic Diffuse Global Illumination
Zihao Liu, Jing Huang, Allan Rocha, Jim Malmros, Jerry Zhang
Open publisher page 0 citations
Abstract
In this paper, we propose a first and efficient ray allocation technique for Dynamic Diffuse Global Illumination (DDGI) using Multiple Importance Sampling (MIS). Our technique, IS-DDGI, extends DDGI by incorporating a set of importance-based ray strategies that analyze, allocate, and manage ray resources on the GPU. We combine these strategies with an adaptive historical and temporal frame-to-frame analysis for an effective reuse of information and a set of GPU-based optimizations for speeding up ray allocation and reducing memory bandwidth. Our IS-DDGI achieves similar visual quality to DDGI with a speedup of 1.27x to 2.47x in total DDGI time and 3.29x to 6.64x in probes ray tracing time over previous technique [Majercik et al. 2021]. Most speedup of IS-DDGI comes from probes ray tracing speedup.