Image credit: Headshots courtesy of NVIDIA Research
NVIDIA Research leaders Sanja Fidler, Aaron Lefohn, and Ming-Yu Liu are presenting the SIGGRAPH 2025 | NVIDIA Research Special Address on Monday, 11 August at 4 pm PDT. Ahead of this session, SIGGRAPH connected with these leaders to preview their presentation and gain insight into how emerging technologies expand possibilities for storytelling, how advances in computer graphics are bettering AI, and more. Read on to prepare for the special address at SIGGRAPH 2025.
SIGGRAPH: SIGGRAPH 2025 explores how new technologies create new opportunities for storytelling. How are emerging technologies like neural rendering and world foundation models expanding the possibilities for immersive storytelling?
Aaron Lefohn (AL): The newest technologies, like neural rendering and world foundation models, are making virtual worlds even more detailed and lifelike while giving creators, researchers, and developers more tools to accelerate their work.
SIGGRAPH: How is the convergence of AI and computer graphics reshaping traditional pipelines in industries like media, automotive, and robotics? What should SIGGRAPH participants keep in mind when considering how their work impacts industries beyond technology?
Ming-Yu Liu (ML): The big takeaway that hopefully everyone at SIGGRAPH walks away with is that AI and graphics are transforming every industry, not just media or content creation.
Sanja Fidler (SF): Exactly. In the past, breakthroughs in industries were oftentimes more siloed — now participants can think about all the ways they can apply their research and work across every field.
SIGGRAPH: We understand that AI is transforming computer graphics — but how are advances in computer graphics also helping AI get smarter?
AL: SIGGRAPH is synonymous with cutting-edge computer graphics, and NVIDIA’s been here a long time presenting our latest and greatest at the conference — RTX and DLSS were both announced at SIGGRAPH.
With the advent of AI, there’s been an exciting feedback loop where AI improves graphics and graphics improves AI, with continuously more impressive results. Right now, we’re embedding differentiable graphics into our AI training to teach AI how to generate 3D worlds from images. And computer graphics is improving AI when it comes to creating high-quality synthetic datasets and realistic simulation environments, which can be used to train and test AI for the real world.
Image credit: Courtesy of NVIDIA Research
SIGGRAPH: What is one key takeaway you hope resonates with conference participants after the SIGGRAPH 2025 | NVIDIA Research Special Address? How does coming together as professionals help move this conversation forward?
ML: This year’s SIGGRAPH is taking place in the age of physical AI. I hope everyone who attends the conference comes away thinking about how graphics, and all the work that researchers around the world have done in that field, are making this moment possible.
SIGGRAPH: Many of your colleagues at NVIDIA are presenting content at SIGGRAPH 2025. After the special address, what other SIGGRAPH 2025 content should conference participants attend and interact with to continue the conversation from your presentation?
SF: I’d recommend checking out Rendering Day on Tuesday, 12 August. Experts in art, design, and rendering technology will be representing the latest advancements in rendering and how it’s enhancing everything from visual effects to design virtualization. Wednesday, 13 August is OpenUSD Day, which anyone interested in adopting OpenUSD for their applications — whether its robotics, industrial AI, or content creation — would benefit from attending. For anyone who enjoys our keynote, I’d also recommend our Course, “An Introduction to Neural Shading”, on Thursday, 14 August.
There’s still time to register for SIGGRAPH 2025! Join us in Vancouver or online, 10–14 August, and secure your spot at the SIGGRAPH 2025 | NVIDIA Research Special Address.
Sanja Fidler is vice president of AI research at NVIDIA, leading the company’s Spatial Intelligence Lab. She is also an associate professor at the University of Toronto, and an affiliate faculty member at the Vector Institute, which she co-founded. Fidler co-authored over 130 scientific papers in the fields of computer vision, machine learning and NLP. She has served as Area Chair for a variety of conferences, including the Conference on Computer Vision and Pattern Recognition (CVPR), International Conference in Computer Vision (ICCV), the Conference on Empirical Methods in Natural Language Processing (EMNLP), the International Conference on Learning Representations (ICLR), the Conference on Neural Information Processing Systems (NeurIPS), and SIGGRAPH.
Fidler has received the NVIDIA Pioneer of AI Award, Amazon Academic Research Award, Facebook Faculty Award, Early Researcher Award, University of Toronto’s Innovation Award, and the Connaught New Researcher Award. In 2018, she was appointed as the Canadian CIFAR AI Chair. She has also been ranked among the top three most influential AI female researchers in Canada by Re-WORK, was named in Globe and Mail’s Changemakers in 2025, and was listed in the top people in AI in 2023 by the Business Insider. With her co-workers, she received the Best Paper Honorable Mention award at CVPR’17, SIGGRAPH 2023, and Best Paper Award at SIGGRAPH Asia 2023. Her main research interests are in spatial intelligence: 3D content creation, spatial understanding, and simulation for robotics.
Aaron Lefohn is vice president of graphics research at NVIDIA, overseeing teams focused on rendering, AI graphics, and graphics systems. His teams’ inventions have played key roles in bringing path tracing to real-time graphics and pioneering real-time AI computer graphics. Recent NVIDIA products derived from his teams’ inventions include DLSS, RTX Path Tracing, RTX Dynamic Illumination, Neural Shading, the Slang shading language, and more.
Aaron has led real-time rendering and graphics programming model research teams for over 15 years and has productized many inventions into games, professional graphics software, GPU hardware, and GPU graphics APIs.
Aaron holds a Ph.D. in computer science from UC Davis and an M.S. in computer science from the University of Utah.
Ming-Yu Liu is a vice president of research at NVIDIA and a fellow of IEEE. He leads the Deep Imagination Research group at NVIDIA, which currently focuses on Generative AI for Physical AI. His research team has helped create several new product categories for NVIDIA, including NVIDIA Cosmos, a developer-first world foundation model platform for Physical AI; NVIDIA Edify, a family of Generative AI models that powers Getty Images and Shutterstock’s GenAI services; NVIDIA Canvas [GauGAN], a real-time painting tool that uses GANs to turn simple brushstrokes into photorealistic images; and NVIDIA Maxine [LivePortrait], an AI-first cloud-native video streaming platform. His research group publishes scientific papers in top-tier AI conferences regularly, including NeurIPS, ICLR, ICML, CVPR, ICCV, ECCV, and SIGGRAPH. Several of their papers have received prestigious awards.
