This is a modal window.
Beginning of dialog window. Escape will cancel and close the window.
End of dialog window.
Training high-performing AI models often relies on real-world data that may be scarce and often expensive to collect and label. Synthetic data generated from custom pipelines built on NVIDIA Omniverse Replicator provides a way to generate training data from 3D simulations for use cases like visual inspection to detect defects on manufactured parts.
In this four-part tutorial, learn how to train a defect detection model with synthetic data using Omniverse Replicator. This second tutorial will walk you through how to efficiently load and manipulate an OpenUSD scene, add textures and generate annotated images of your CAD part with defects.
Get started with synthetic data generation on NVIDIA Omniverse: https://developer.nvidia.com/omniverse/replicator
Watch the full four-part tutorial here: https://www.nvidia.com/en-us/on-demand/playlist/playList-35d98b97-8abf-4f92-883a-c898801f28b4/