AI in Action: NVIDIA Showcases New Research, Enhanced Tools for Creators at SIGGRAPH

The future of graphics is here, and AI is leading the way.

At the SIGGRAPH 2020 virtual conference, NVIDIA is showcasing advanced AI technologies that allow artists to elevate storytelling and create stunning, photorealistic environments like never before.

NVIDIA tools and software are behind the many AI-enhanced features being added to creative tools and applications, powering denoising capabilities, accelerating 8K editing workflows, enhancing material creation and more.

Get an exclusive look at some of our most exciting work, including the new NanoVDB library that boosts workflows for visual effects. And check out our groundbreaking research, AI-powered demos, and speaking sessions to explore the newest possibilities in real-time ray tracing and AI.

NVIDIA Extends OpenVDB with New NanoVDB

OpenVDB is the industry-standard library used by VFX studios for simulating water, fire, smoke, clouds and other effects. As part of its collaborative effort to advance open source software in the motion picture and media industries, the Academy Software Foundation (ASWF) recently announced GPU-acceleration in OpenVDB with the new NanoVDB for faster performance and easier development.

OpenVDB provides a hierarchical data structure and related functions to help with calculating volumetric effects in graphic applications. NanoVDB adds GPU support for the native VDB data structure, which is the foundation for OpenVDB.

With NanoVBD, users can leverage GPUs to accelerate workflows such as ray tracing, filtering and collision detection while maintaining compatibility with OpenVDB. NanoVDB is a bridge between an existing OpenVDB workflow and GPU-accelerated rendering or simulation involving static sparse volumes.

Hear what some partners have been saying about NanoVDB.

“With NanoVDB being added to the upcoming Houdini 18.5 release, we’ve moved the static collisions of our Vellum Solver and the sourcing of our Pyro Solver over to the GPU, giving artists the performance and more fluid experience they crave,” said Jeff Lait, senior mathematician at SideFX.

“ILM has been an early adopter of GPU technology in simulating and rendering dense volumes,” said Dan Bailey, senior software engineer at ILM. “We are excited that the ASWF is going to be the custodian of NanoVDB and now that it offers an efficient sparse volume implementation on the GPU. We can’t wait to try this out in production.”

“After spending just a few days integrating NanoVDB into an unoptimized ray marching prototype of our next generation renderer, it still delivered an order of magnitude improvement on the GPU versus our current CPU-based RenderMan/RIS OpenVDB reference,” said Julian Fong, principal software engineer at Pixar. “We anticipate that NanoVDB will be part of the GPU-acceleration pipeline in our next generation multi-device renderer, RenderMan XPU.”

Learn more about NanoVDB.

Research Takes the Spotlight

During the SIGGRAPH conference, NVIDIA Research and collaborators will share advanced techniques in real-time ray tracing, along with other breakthroughs in graphics and design.

Learn about a new algorithm that allows artists to efficiently render direct lighting from millions of dynamic light sources. Explore a new world of color through nonlinear color triads, which are an extension of gradients that enable artists to enhance image editing and compression.

And hear from leading experts across the industry as they share insights about the future of design:

Omniverse: Get the latest updates on NVIDIA Omniverse, a computer graphics and simulation platform that lets artists work seamlessly in real time across popular content-creation applications.
RTX-accelerated ray tracing with OptiX 7: Learn about the general concepts behind OptiX and how to fully optimize your application to get the most out of hardware ray-tracing cores of modern GPUs.
Making Machine Learning Work: From Ideas to Production Tools: Experts from NVIDIA, Weta Digital, Ubisoft and more will discuss how machine learning is enhancing the production of computer graphics.
Join a virtual round table with AWS or visit our virtual booth with Microsoft to learn more about Quadro virtual workstations and customer cases.

Check out all the groundbreaking research and presentations from NVIDIA.

Eye-Catching Demos You Can’t Miss

This year at SIGGRAPH, NVIDIA demos will showcase how AI-enhanced tools and GPU-powered simulations are leading a new era of content creation:

Synthesized high-resolution images with StyleGAN2: Developed by NVIDIA Research, StyleGAN uses transfer learning to produce portraits in a variety of painting styles.
Mars lander simulation: A high-resolution simulation of retropropulsion is used by NASA scientists to plan how to control the speed and orientation of vehicles under different landing conditions.
AI denoising on Blender: RTX AI features like OptiX Denoiser enhances rendering to deliver an interactive ray-tracing experience.
8K video editing on RTX Studio laptops: GPU acceleration for advanced video editing and visual effects, including AI-based features in DaVinci Resolve, helping editors produce high-quality video and iterate faster.