15

GitHub - grantwilk/DNOISE

 5 years ago
source link: https://github.com/grantwilk/DNOISE
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.

README.md

alt text

D-NOISE, powered by NVIDIA’s OptiX AI-Acclerated Denoiser, is a game-changing denoising platform that rapidly accelerates the process of rendering high-quality noiseless images in Blender.

Download:

Download the add-on: https://remingtongraphics.net/tools/d-noise/

Features:

Clean Implementation: A clean and orderly implementation makes D-NOISE more convenient than the default Cycles denoiser. Easily access the denoiser from the UV/Image editor or the render layers tab of the properties panel. alt text alt text Takes Seconds: GPU-acceleration enables the underlying AI denoiser to accurately eliminate noise from your renders in no more than a few seconds. Denoise Anything: D-NOISE isn’t exclusive to renders like the Cycles denoiser is. D-NOISE can run on any image loaded into the UV/Image Editor including texture bakes and even photos! alt text alt text HDR Training: HDR training data enables the underlying AI denoiser to maintain detail in nearly any lighting scenario, even harsh darkness.

Installation

Installation guide: https://remingtongraphics.net/tools/d-noise/

Documentation

Documentation: https://remingtongraphics.net/tools/d-noise/

System Requirements

  • Blender 2.7x (not yet compatible with 2.80)
  • Windows 7, 8, or 10
  • NVIDIA GTX 600 Series or newer
  • GeForce driver version 390.xx or newer

Acknowledgements

Special thanks to Declan Russell for sharing his standalone NVIDIA AI Denoiser and Deep Blender for sharing his normal conversion script. Also, massive thanks to everyone who helped in beta testing!


report

Recommend

About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK