Parallel Domain Empowers Customers to Construct Synthetic Datasets Using Generative AI via Their API
Parallel Domain Empowers Customers with Generative AI-Driven API for Synthetic Dataset Generation
chandni
San Francisco-based startup, Parallel Domain, has unveiled its latest API called Data Lab, which allows customers to generate synthetic datasets using powerful generative AI technology. By leveraging dynamic virtual worlds and simulating diverse scenarios, machine-learning engineers now have the ability to create datasets tailored to their specific needs.
Kevin McNamara, Founder and CEO of Parallel Domain, explained that accessing the API is as simple as installing it from GitHub and using Python code to generate datasets. With Data Lab, engineers can now generate objects that were previously unavailable in the startup's asset library. By utilizing 3D simulation, the API provides a solid foundation for engineers to overlay real-world elements, introducing randomness and complexity. For instance, engineers can train their models to navigate a highway with a flipped-over cab obstructing two lanes or teach a robotaxi to identify a person dressed in an inflatable dinosaur costume.
The primary aim of this endeavor is to grant autonomy, drone, and robotics companies greater control and flexibility in generating and manipulating synthetic datasets.
