← All resources

Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling

Auto-discovered

Paper Advanced

Pengpei Hong, Meng Duan, Beibei Wang, Cem Yuksel, Tizian Zeltner, Daqi Lin

Visit resource ↗

Description

Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling presents a Utah Graphics Lab method by Pengpei Hong, Meng Duan, Beibei Wang, Cem Yuksel, Tizian Zeltner, and Daqi Lin that speeds Specular Manifold Sampling for interactive caustics via sample-space partitioning and ReSTIR.

It covers costly SMS Newton iterations under tight budgets, tile-based partitioning that bounds manifold walks and builds per-frame priors, and spatiotemporal reuse that amortizes sample generation. Timed comparisons against path tracing and SMS appear with an interactive result viewer.

It is for advanced researchers and engineers working on Monte Carlo caustics and real-time path reuse. Familiarity with SMS and ReSTIR is expected.

Access is a free project page with abstract, comparisons, scene thumbnails, and notes on a Falcor 8.0 PSMSReSTIR module.

It is useful when you need lower-variance interactive caustics than plain SMS in similar millisecond budgets. Link last verified accessible on 2026-09-29. Link last verified accessible on 2026-09-28.