This publication describes techniques for grip suppression of a touchscreen display of a computing device using a machine-learned technique. When a user intentionally or unintentionally touches the display (e.g., with a hand or a finger), a Touch Manager of the computing device performs operations to determine a user intent associated with the touch input to prevent false triggering of the display (e.g., by a grip of a hand holding the device). A machine-learned (ML) model calculates the likeliness of an intentional touch input (e.g., a tap, a swipe, or a scroll of a hand or a finger to input or manage information on the device) by identifying and assigning weights to features of the touch input. A total weight is calculated for each touch input and compared to a default threshold (e.g., an accepted threshold associated with an intentional touch input), which may be adjusted to ensure accuracy of user-intent predictions. After the Touch Manager verifies the user-intent predictions using heuristic and/or hysteresis logic, the computing device may perform operations to suppress or enable the touch input.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Kuan, Katherine and Chang, Mark, "Grip Suppression using a Machine-Learned Technique", Technical Disclosure Commons, (January 24, 2021)