US 12,346,516 B1
Touch input force estimation using machine learning
Yael Livne, Tel Aviv (IL); Adam Hakim, Tel Aviv (IL); and Nadav Linenberg, Even Yehuda (IL)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Jan. 2, 2024, as Appl. No. 18/402,316.
Int. Cl. G06F 3/041 (2006.01); G06F 3/044 (2006.01); G06N 20/00 (2019.01)
CPC G06F 3/0416 (2013.01) [G06F 3/044 (2013.01); G06N 20/00 (2019.01); G06F 2203/04105 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
a touch interface including a plurality of touch sensors, the touch interface being configured to output a touch heatmap based at least on touch input detected by the plurality of touch sensors, the touch heatmap including a plurality of capacitance values corresponding to the touch input detected by the plurality of touch sensors;
a logic subsystem; and
a storage subsystem holding instructions executable by the logic subsystem to:
execute a machine-learning model configured to receive the touch heatmap and output a force estimation of the touch input detected by the plurality of touch sensors based at least on analyzing the touch heatmap, wherein the machine-learning model is trained based at least on training data generated by a training touch interface including a plurality of force sensors spatially distributed across the training touch interface, wherein the training data includes a plurality of training touch heatmaps representing a plurality of different instances of training touch input to the training touch interface and a corresponding set of force values output by the plurality of force sensors based at least on the plurality of different instances of training touch input; and
execute a computing operation based at least on the force estimation output from the machine-learning model.