AI application in 10 minutes

For this demo, we will create an AI model in a short time and run it on an embedded system. With our own AI flow, we can quickly and efficiently create prototypes for AI applications. Our flow consists of the following steps, which may or may not be followed depending on the application.

The first step is generating data. Data is essential because the quality of a model is closely linked to the quality of the data. Since obtaining data is not always possible, or is time-consuming and costly, we have developed a tool that allows us to generate synthetic data ourselves.

The second step is training the model. For this, in many cases, we use existing pre-trained models, which we retrain with our data for a fast and efficient result.

The final step is integrating this model into an embedded system. For this, we again make use of our own libraries and tooling, allowing us to quickly create a custom application that is target-independent.

Exhibitor: Core-Vision