Extend Darcy AI for your use case

Darcy CLoud Project Page

Darcy AI can be extended to leverage different input data sources, use custom AI models , and output data anywhere you want.

To add new input data sources

You can use input stream code files from any existing source or you can create your own. As an example, you can add an Input Stream to your application that adds audio, thermal vision, LiDAR, or something more simple like data from a GPS receiver that is updated every 10 seconds.

Example Input Stream

To add new AI models

You can use existing Perceptors provided by others or you can create your own. Perceptors are the AI processing units in Darcy AI. You can include any AI model in a Perceptor file and it will become available to use in your Darcy AI pipeline in your application. As an example, you can add a Perceptor that identifies cars and finds their license plates and then another Perceptor that performs optical character recognition on the license plates that have been identified.

Example Perceptor

To change how Darcy AI outputs data

You can use output stream code files from any existing source or you can create your own. As an example, you can add an Output Stream to your application that stores data in an AWS S3 bucket or an Output Stream that turns the fully annotated content into an RTSP network video feed.

Example Output Stream

More Information and example code