A computing device (e.g., a smartphone, a laptop computer, a tablet computer, a smartwatch, etc.) may use a machine learning model to classify user inputs as a back gesture for navigating with respect to graphical user interfaces (GUI) of the computing device. The computing device may apply a machine learning model to input data associated with the user input (e.g., (x,y) coordinates of the user inputs) and a context of the computing device (e.g., an application (“app”) that is currently executing on the computing device, the width of the computing device, the orientation of the computing device, etc.) to determine a degree of likelihood of the user input being a back gesture. If the degree of likelihood of the user input being a back gesture satisfies a threshold, the computing device may execute a back action associated with the back gesture. If the degree of likelihood does not satisfy the threshold, the computing device may execute a different action or may discard the user input. The machine learning model may be trained on a computing system (e.g., a remote server) distinct from the computing device while the trained machine learning model may be stored at the computing device.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Karimzadehgan, Maryam and Upstill, Trystan, "AI-ASSISTED GESTURE NAVIGATION FOR COMPUTING DEVICES", Technical Disclosure Commons, (May 03, 2021)