Automate 3D Pipelines With AI Agents and Vision-Language Models (VLMs)

OpenUSD developers already have assets, review tools, and content pipelines. This livestream explores how AI agents and vision-language models (VLMs) can automate the repetitive work around 3D asset preparation without rebuilding existing pipelines. You’ll see NVIDIA OpenUSD content agents analyze USD assets, classify physics properties, assign materials, and generate textures. The session will also showcase these agents running inside a lightweight OpenUSD viewer and renderer web UI built with Codex and connected to ovrtx, demonstrating how developers can prototype AI-assisted tools that bring intelligent automation directly into OpenUSD workflows and web-based 3D experiences. You will learn: How AI agents and VLMs can inspect USD assets and reason over 3D scene data How content agents can automate material assignment, physics property classification, and texture generation How to prototype AI-assisted OpenUSD tools in a lightweight web UI connected to NVIDIA Omniverse rendering workflows How developers can bring intelligent automation directly into OpenUSD pipelines and web-based 3D experiences How to use AI agents to help streamline your 3D content workflows 📆 Check out the full calendar for all of our upcoming events → https://nvda.ws/3JqaWnA ---------------------------------------------------------------------------- ⬇️Get Started → https://nvda.ws/4cZAZO1 👀Explore OpenUSD → https://nvda.ws/3CeozBQ 👥Join the Community → https://nvda.ws/3ZMfc6e