In this episode of the NVIDIA AI Podcast, host Noah Kravitz interviews Aaron Luk, NVIDIA's Director of Product Management for Simulation Technology, about OpenUSD (Universal Scene Description) and its impact on 3D graphics, simulation, and physical AI. They discuss the benefits of OpenUSD, including its ability to unify different data sources, facilitate collaboration, and accelerate the development of physical AI by creating realistic training environments for robots and autonomous systems. Aaron explains how OpenUSD helps bridge the sim-to-real gap, supports the creation of digital twins for industrial use cases, and enables the generation of synthetic data. He also touches on the standardization efforts within the OpenUSD ecosystem and provides resources for listeners interested in getting started with OpenUSD, such as the NVIDIA Deep Learning Institute and the OpenUSD GitHub repository.
Sign in to continue reading, translating and more.
Continue