NVIDIA at SIGGRAPH 2024
Fireside Chat With NVIDIA CEO Jensen Huang
Monday, July 29 | 2:30 p.m. MT
NVIDIA Founder and CEO Jensen Huang sits down with Lauren Goode, senior writer at WIRED, to discuss how accelerated computing and generative AI are transforming industries and creating new opportunities for innovation and growth.
The session will also be live-streamed.
Featured Programs
Enjoy a complimentary expo pass on NVIDIA using code NVI4327 when you start the registration process here.
Schedule at a Glance
-
Sun, July 28
-
Mon, July 29
-
Tue, July 30
-
Wed, July 31
-
Thu, Aug. 1
| 8:00–10:00 a.m. MT | Generative AI for Graphics, Vision and 3DBirds of a Feather | Room 708-710 |
| 9:00–10:30 a.m. MT | On Smoothly Varying Frame Timing in First-Person GamingTalk | Mile High 2 |
| 9:00–10:30 a.m. MT | Studying Esports Competition: Piloting Methodology for User Studies During TournamentsTalk | Mile High 2 |
| 9:00 a.m.–12:15 p.m. MT | Warp: Differentiable Spatial Computing for PythonCourse | Mile High 1A |
| 9:00–10:30 a.m. MT | Diffusion Texture PaintingTechnical Paper | Mile High 2 |
| 10:45 a.m.–12:15 p.m. MT | A Differential Monte Carlo Solver for the Poisson EquationTechnical Paper | Mile High 4 |
| 2:00–3:30 p.m. MT | Path-space Differentiable Rendering of Implicit SurfacesTechnical Paper | Mile High 4 |
| 2:30–3:30 p.m. MT | NVIDIA CEO Fireside Chat: AI and Graphics for the New Computing RevolutionFireside Chat | Bluebird Ballroom - Level 3 |
OpenUSD Day
Explore OpenUSD Day sessions hosted by industry leaders or check out partner talks NVIDIA is participating in.
Generative AI Day
Explore Generative AI Day sessions hosted by industry leaders or check out partner talks NVIDIA is participating in.
| 8:00–9:00 a.m. MT | Immersive Visualisation for Science, Research, Art and Metaverse ApplicationsBirds of a Feather | Room 708-710 |
| 9:00–10:30 a.m. MT | fVDB: A Deep-Learning Framework for Sparse, Large-Scale, and High-Performance Spatial IntelligenceTechnical Paper | Mile High 2 |
| 10:45 a.m.–12:15 p.m. MT | Motion-I2V: Consistent and Controllable Image-to-video Generation With Explicit Motion ModelingTechnical Paper | Mile High 1A |
| 2:00–3:30 p.m. MT | Interactive Character Control With Auto-regressive Motion Diffusion ModelsTechnical Paper | Mile High 4 |
| 2:00–3:30 p.m. MT | SuperPADL: Scaling Language-Directed Physics-Based Control With Progressive Supervised DistillationTechnical Paper | Mile High 4 |
| 2:00–3:30 p.m. MT | Surface-filling Curve Flows via Implicit Medial AxesTechnical Paper | Mile High 2 |
NVIDIA Training Labs at SIGGRAPH
Develop your skills with training in AI, research, OpenUSD, and more by joining our hands-on labs, complimentary for the Full Conference and Experience attendees.
Monday, July 29
10:00–11:30 a.m. MT
Encore Session: 12:30–2:00 p.m. MT
Explore the future of retail with spatial computing. You'll have the opportunity to create an application for the Apple Vision Pro (AVP) that allows users to configure a photoreal, 3D retail asset. This includes developing an application and setting it up to communicate with a product configurator built with the Omniverse Kit SDK and OpenUSD. You'll also learn how to implement custom Swift UI to interact with the virtual product in real time. By the end of this training, participants will be able to leverage advanced technologies to create immersive, interactive retail experiences.
Monday, July 29
9:00–11:00 a.m. MT
Foundation models (FMs) are pretrained AI models with vast potential for various natural language processing tasks, from language translation to sentiment analysis. Join this training lab, where we’ll explore the basics of foundation models, their significance, applications, and the latest developments in the field of AI.
- Learn the evolution of foundation models—tracing their path from the early days of the Perceptron to their current role in shaping AI, like ChatGPT.
- Discuss the challenges foundation models face in today’s AI landscape.
- Build custom AI models and applications that leverage foundation models.
Monday, July 29
12:00–2:00 p.m. MT
Semantic retrieval has become a common tool for helping to drive large language models toward data-grounded reasoning. However, naive techniques start to break down when non-textual inputs come into the picture. Instead of ignoring non-textual information, we’ll tackle multimodal document ingestion and retrieval.
- Construct the simple retrieval-augmented generation pipeline for context enrichment.
- Consider approaches for reasoning with images to help an LLM-powered agent converse about image-dense research papers.
Tuesday, July 30
9:00–11:00 a.m. MT
As AI continues to make its way into popular use, deep learning components from a wide range of disciplines are quickly getting incorporated into end-user applications, and emerging workflows that combine multiple seemingly disjoint models or input types are able to operate surprisingly well.
- Investigate the primitives that help to make up anything from MNIST image generators to vision-guided language models.
- Discuss large model orchestration and show how they can be used to construct end-to-end multimodal agent systems.
Tuesday, July 30
11:00 a.m.–1:00 p.m. MT
Interactive chatbots are gaining popularity in use across multiple industries, such as virtual assistants, customer service agents, sales support agents, and other applications. This training lab covers important topics to help users understand and implement their custom Digital Human.
- Discuss the key benefits of enhancing chatbots toward interactive Digital Humans.
- Detail the key components of a Digital Human.
- Build and deploy an AI-powered Digital Human.
Tuesday, July 30
1:00–5:00 p.m. MT
Join us anytime during this drop-in workshop to learn about OpenUSD foundations, supported by our USD Code NIM. Whether you're new to OpenUSD or looking to enhance your skills, we encourage you to check out these self-guided courses. You'll learn the foundational concepts of OpenUSD, acquire basic Python programming skills relevant to OpenUSD, gain hands-on experience through practical exercises and real-world examples, and use USD Code to get answers and guidance as you progress.
*Self-paced
Wednesday, July 31
9:00–10:30 a.m. MT
This course on composition arcs is designed for developers and 3D artists looking to use OpenUSD’s powerful composition capabilities to create modular, scalable, and efficient 3D scenes. We'll cover the fundamental concepts of composition arcs, layers and layer stacks, references and payloads, variant sets, and debugging and introspection. By the end of this course, you'll have the skills to build and manage complex 3D scenes with ease.
Wednesday, July 31
11:00 a.m.–12:30 p.m. MT
In this course, we’ll cover the importance of asset structures in production environments, model hierarchies, and how to organize work streams with layer stacks and streamline collaborative workflows while optimizing performance. Anyone working with OpenUSD will be able to create scalable, modular, and high-performance scenes that facilitate collaboration, content reuse, and parallel work streams in any industry. This includes everything from manufacturing and digital twins, to production and animation, to retail and product design.
Wednesday, July 31
1:30–3:00 p.m. MT
Unlock the full potential of your 3D workflows with this comprehensive course on instancing with OpenUSD, designed for developers and 3D artists. Whether you're building OpenUSD pipelines for digital twins, media and entertainment, AEC, or XR, you'll learn essential best practices and advanced techniques to enhance your workflows, optimize performance, and achieve scalable, high-quality results.
Wednesday, July 31
3:30–5:00 p.m. MT
Learn to make the most of your 3D workflows and development pipelines by joining us for a course on data exchange with OpenUSD. This course is designed for developers and 3D artists looking to optimize their existing 3D workflows by mastering the process of exchanging data between OpenUSD and other data formats. We’ll review the principles of conceptual data mapping, creating OpenUSD conversion scripts with Python, and going through asset validation and testing. By the end of this course, you'll be able to implement and manage data exchange processes between OpenUSD and your favorite DCC tools.
NVIDIA Research Presentations at SIGGRAPH
All in on AI
NVIDIA Programs
-
Researchers
-
Educators
-
Developers
-
Creators
-
Startups
-
Designers
Like No Place You’ve Ever Worked
At NVIDIA, you’ll solve some of the world’s hardest problems and discover never-before-seen ways to improve the quality of life for people everywhere. From healthcare to robots, self-driving cars to blockbuster movies, the list of new opportunities is growing every single day. Become an NVIDIAN today.
Want to hear what else we have in store at SIGGRAPH? Sign up to get notified each time we update this page with speakers, schedule, demos, and more.