NVIDIA announced the NVIDIA Cosmos Data Factory (CDF), an open platform for physical AI datasets. It provides five cloud-native blueprints and six composable skills to accelerate data curation, processing, and synthetic data generation. The CDF Orchestration core skill coordinates workflow steps, deploys required services, and manages execution on a cluster for all five reference blueprints.
What happened
NVIDIA announced the NVIDIA Cosmos Data Factory (CDF), an open platform for physical AI datasets. It provides five cloud-native blueprints and six composable skills to accelerate data curation, processing, and synthetic data generation. The CDF Orchestration core skill coordinates workflow steps, deploys required services, and manages execution on a cluster for all five reference blueprints. Five training data use cases include generating driving scenarios for autonomous vehicle training, augmenting image attributes for search and re-identification, augmenting videos across weather and lighting, generating rare event videos, and generating defect images for visual inspection. NVIDIA Cosmos Data Factory is expected to shorten AI regression test cycles by up to 85% and enable going to market in under two weeks compared to an estimated six months.
What to weigh
- NVIDIA Cosmos Data Factory is expected to shorten AI regression test cycles by up to 85%
- Going to market in under two weeks compared to an estimated six months
FAQ
What are the five cloud-native blueprints?
NVIDIA Cosmos Data Factory (CDF) provides five cloud-native blueprints and six composable skills to accelerate data curation, processing, and synthetic data generation for physical AI.
What are the six composable skills?
NVIDIA Cosmos Data Factory (CDF) provides five cloud-native blueprints and six composable skills to accelerate data curation, processing, and synthetic data generation for physical AI.
How many training data use cases are there?
Five training data use cases include generating driving scenarios for autonomous vehicle training, augmenting image attributes for search and re-identification, augmenting videos across weather and lighting, generating rare event videos, and generating defect images for visual inspection.
Source: NVIDIA
