The company’s Research and Engineering leaders Neil Ashton, Ming-Yu Liu and Edward Liu will talk on advances in neural rendering, world models and AI-driven simulation next Monday, July 20; conference runs July 19-23 in Los Angeles.
NVIDIA Research and Engineering leaders Neil Ashton, Ming-Yu Liu and Edward Liu will present the sponsored keynote “Next Era of Graphics — Neural Rendering, World Models, and Simulation” at SIGGRAPH 2026 on Monday, July 20, at 3:45 p.m. PDT. SIGGRAPH 2026 runs July 19-23, with the exhibition taking place July 21-23.
The presentation will explore recent advances shaping computer graphics and simulation, including breakthrough neural rendering techniques, progress in developing world models and simulation methods for AI, built by AI. Through recent research examples and demos, the speakers will examine how these technologies are changing the way digital worlds are built, rendered and used across creative tools, industrial design, robotics and autonomous systems.
Other NVIDIA sessions include:
- Career Growth in Physical AI: Bringing 3D Graphics Skills Into Robotics and Digital Twins
- Building Physics-Aware, RTX-Enabled Apps With NVIDIA Omniverse™ Libraries and AI Agents
- OpenUSD Proposals Across Domains: Challenges and Methodology
- OpenUSD for AI Agents, Model-Driven Workflows, and 3D Graphics Pipelines
- Building a GPU Production Renderer With OptiX™ at Netflix Animation Studios
- Real-Time Short-Form Content: Behind the Scenes of the ILM Minis
- How AI Is Changing Education
Ashton is a Distinguished Engineer at NVIDIA focused on AI physics and agentic AI for computational engineering. A Fellow of the Institution of Mechanical Engineers, he works across accelerated simulation, scientific machine learning and physical AI, helping bring advanced AI methods into practical aerospace, automotive and industrial workflows. He pioneered the first open datasets for complete aircraft, HiLiftAeroML, and road-vehicle configurations, DrivAerML, and leads international workshops including AutoCFD and HLPW-6. His current work explores AI agents, physics foundation models, surrogate modeling and accelerated digital twins, with a focus on systems that can reason about, simulate and optimize complex physical designs. Before NVIDIA, he held positions at AWS, NASA, Formula 1 and the University of Oxford.
Liu is an IEEE Fellow who leads the NVIDIA Cosmos Lab, previously the Deep Imagination Research group, focusing on generative AI for Physical AI, including world foundation models that enable machines to simulate, reason about and interact with the physical world. Driven by the vision of building a physics-grounded “Matrix” for robots, Liu bridges research with real-world platforms used by millions, with his team's work powering NVIDIA technologies including Cosmos, Edify, Canvas and Maxine, along with contributions to text-to-image, text-to-video and text-to-3D models. He previously worked at Mitsubishi Electric Research Labs and earned his Ph.D. from the University of Maryland, College Park.
Edward Liu is Director of Applied Deep Learning Research at NVIDIA, where he leads research in AI for real-time graphics and generative visual computing. He has helped transform real-time graphics through AI, leading advances across the DLSS technology suite, which is widely adopted across the industry and has become a defining RTX product differentiator. His current work spans DLSS 5 Neural Rendering, real-time visual generation and broader research shaping the future of AI-powered interactive visual computing.
Other key events and happenings during the conference:
The Computer Animation Festival includes screenings of the Electronic Theater and Animation Theater, giving attendees multiple ways to experience emerging animation trends from around the world. Several top films will also be featured in the SIGGRAPH Computer Animation Festival Traveling Show, a touring collection that brings the Electronic Theater to campuses, festivals and communities worldwide.
The Production Sessions program will offer behind-the-scenes looks at films, immersive experiences and scientific visualization projects. In “Wētā FX Presents: Avatar: Fire and Ash,” Wētā FX and Lightstorm Entertainment will examine more than 3,100 shots created for the return to Pandora, including the blending of extensive native stereo photography with fully digital, performance-captured characters.
The Wētā FX and Lightstorm Entertainment session will also cover how the team handcrafted elements of the world, including creatures, clans and the film's titular fire, created in Kora, a new addition to Wētā FX's global simulation framework, Loki. The team created more than 2,500 facial motion shots for 30 speaking characters using a proprietary facial system designed to preserve each actor's performance.
Additional Production Sessions include “Visualizing the Moon for Artemis II,” in which NASA shows how cinematic scientific visualization built from publicly available datasets and high-end 3D rendering recreates the exact lighting astronauts will encounter on the far side of the Moon. The work has helped plan observations and train the Artemis II crew for the first crewed flight to the Moon since Apollo 17.
Pixar Animation Studios will present “We're All In This Together: The Making of Disney and Pixar's Hoppers,” focusing on the cross-departmental pipeline that allowed physical simulation and artistically directed stylization to coexist in every shot while balancing broad comedy, suspense and the beauty of nature photography.
Industrial Light & Magic will present “Bringing The Mandalorian and Grogu to Life: Virtual Production, Creatures, and VFX Innovations,” detailing how real-time technology, virtual production workflows and established visual effects techniques were used to bring the latest “Star Wars” feature to the big screen.
Google DeepMind animation veterans and researchers will present Dear Upstairs Neighbors: Artist-Centered, AI-Assisted Expressionistic Animation, covering how artist-driven, AI-assisted workflows shaped an expressionistic, painterly short, with artists in control as both creators and co-designers throughout the process.
“Production Sessions cover the full production, from the very start of the concept art and the story all the way to the end, giving you a really great general view of the entire pipeline,” SIGGRAPH 2026 Production Sessions chair Valerie Bernard said. “We get to learn about so many creative challenges that teams had to solve by getting multiple departments to work together. They are for everyone. When I first started going to SIGGRAPH, I learned the most about production workflows from Production Sessions, and they are just as amazing for veterans of the industry who like to be reminded why we do this job.”
SIGGRAPH 2026 is the 53rd annual conference hosted by ACM SIGGRAPH. ACM, the Association for Computing Machinery, is an educational and scientific computing society, and ACM SIGGRAPH is its special interest group focused on research, technology and applications in computer graphics and interactive techniques.
Source: SIGGRAPH 2026
Dan Sarto is Publisher and Editor-in-Chief of Animation World Network.
