High accuracy gesture detection for small devices is one of the exciting applications that’s recently become a reality. In action, gesture detection allows for significantly smoother, safer and more intuitive user experiences.
The Imagimob Gesture Detection Library can detect hand gestures using raw radar data from the Acconeer XM122 IoT Module as input. The data is challenging, and for each user there are large data variations. To account for these variations, the library has been trained on gestures performed by several different people. The radar data picks up detailed information which allows for very fine gestures to be recognized. For this to work in a robust way, high requirements have been placed on the accuracy of the AI model.