Background and testing of Imagimob Ready Models
The Imagimob Ready Models are complete machine learning (ML) solutions that are “guaranteed to be robust, high-performing, and production-ready for edge devices”. Ready Models can easily be deployed into existing microcontroller (MCU) hardware, including PSoC 6, without the need for custom development.
To help ensure the “robustness” of the company’s models, they make a list of different scenarios that may be encountered, then test the scenarios. The company also states that they test models in various scenarios around the world to make sure they work with “no bias based on specific geographies or ethnicities”. Imagimob also does field testing on device to test and document realistic performance of the models in an expected hardware setup.
The company is releasing four audio-based models which include Baby Cry for baby monitors and Siren Detection for pedestrians. It also has Coughing Detection and Snoring Detection models which are available as wearable devices in the medtech and health sectors. Imagimob is also working on models within Audio, Radar, IMU and Capacitive Sensing.
Simplifying ML for new companies
Imagimob notes that before they developed these models it was difficult for companies to upgrade products with smart AI features. Not only did ML models have to be custom-made but they needed the right software engineering and AI knowledge as well as a lot of time, money, and resources being spent for the development process.
While the new Imagimob Ready Models “require zero or limited engineering and AI competence to implement. And with all of the development and testing work already put in, they essentially offer a shortcut to the marketplace”.