Simulation and Synthetic Data Generation

This page is for version Dev. For other versions, use one of these documentation indexes: - 3.3.0 (default): https://docs.nvidia.com/vss/3.3.0/llms.txt - Dev: https://docs.nvidia.com/vss/dev/llms.txt

Simulation and synthetic data generation (SDG) are increasingly central techniques to augment the dataset of perception models and to provide a digital twin for end-to-end development & validation of applications.

The goal of SDG is to generate a flexible number of well-placed cameras that provide full scene coverage, along with corresponding ground truth data, to support Multi-Camera Multi-Target Tracking (MTMC) use cases. Detailed dataset introduction can be found in our released PhysicalAI-SmartSpaces.

Briefly, you would need to do the following steps to make use of synthetic data in MTMC use cases.

Follow Quick Start for SDG to download the separate Warehouse USD scene resource (nvidia/vss-warehouse/vss-warehouse-usd-files:3.2.0) and run it in Isaac Sim. The Warehouse app-data archive contains videos and playback data, not USD scenes.