Pith. sign in

REVIEW 1 cited by

Importance Sampling of Many Lights with Reinforcement Lightcuts Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1911.10217 v3 pith:CUB6T7FS submitted 2019-11-22 cs.GR cs.DS

classification cs.GRcs.DS
keywords samplingtechniquedistributionsimportancelearninglightsmanyachieved
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In this manuscript, we introduce a novel technique for sampling and integrating direct illumination in the presence of many lights. Unlike previous work, the presented technique importance samples the product distribution of radiance and visibility while using bounded memory footprint and very low sampling overhead. This is achieved by learning a compact approximation of the target distributions over both space and time, allowing to reuse and adapt the learnt distributions both spatially, within a frame, and temporally, across multiple frames. Finally, the technique is amenable to massive parallelization on GPUs and suitable for both offline and real-time rendering.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Neural Visibility Cache for Real-Time Light Sampling

    cs.GR 2025-06 conditional novelty 7.0 of 10

    A real-time renderer learns a visibility cache with a small neural network and uses it to sample lights, lowering noise versus ReSTIR at comparable cost on tested scenes.

Pith tools