Temporally Adaptive Shading Reuse for Real-time Rendering and Virtual Reality
Author: JOERG H. MUELLER, THOMAS NEFF, PHILIP VOGLREITER, MARKUS STEINBERGER, DIETER SCHMALSTIEG
Institution: Graz University of Technology, Austria
Link: https://dl.acm.org/doi/10.1145/3446790
AbstractMotivation
Temporal coherence has the potential to enable a huge reduction of shading costs in rendering
Current Work
Existing techniques focus either only on spatial shading reuse or cannot adaptively choose temporal shading frequencies
Idea
Temporal shading reuse is possible for extende ...
GPU Accelerated Path Tracing of Massive Scenes
Author: MILAN JAROŠ, LUBOMÍR ŘÍHA, PETR STRAKOŠ, and MATĚJ ŠPEŤKO
Institution: IT4Innovations, VSB–Technical University of Ostrava, Czech Republic
Link: https://dl.acm.org/doi/10.1145/3447807
1. IntroductionGPUs Rendering vs CPUs Rendering
Limited memory size
Example: Pixar’s Coco movie were using up to 120GB, do not fit into the memory of a single GPU
Main contribution
A solution to GPUs rendering memory size limitation
Based on replication of a small amount of scene data, between 1% ~ ...
ExtraNet: Real-time Extrapolated Rendering for Low-latency Temporal Supersampling
AbstractReal-time Rendering
Frame rate
Latency
Spatial Super Sampling: DLSS
Decreasing the rendering time of each frame by rendering at a lower resolution
Temporal Super Sampling
Producing more frames on the fly
Problems:
It’s own computational cost
The latency introduced by interpolating frames from the future
ExtraNet
An efficient neural network that predicts accurate shading results on an extrapolated frame, to minimize both the performance overhead and the latency
Rendered auxiliar ...