Abstract
Proximity sensors used by mobile devices such as a smartphone to activate the screen can malfunction under certain situations, e.g., direct sunlight, holding the phone against the ear at certain angles, etc. This disclosure describes a stable and accurate face proximity detector based on an on-device machine-learning (ML) model that fuses multi-sensor inputs, including inputs from sensors such as accelerometer, gyroscope, ambient light sensor (ALS), proximity sensor, etc., to determine the phone-to-face distance and to correctly activate the screen at a high detection rate and a low false alarm rate.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Kuiper, Chris; Agarwal, Vishal; Tan, Hong Z.; Centazzo, Alessio; and Chen, Keyu, "Selective Mobile Device Screen Activation Using Sensor Fusion and Machine Learning", Technical Disclosure Commons, (January 29, 2025)
https://www.tdcommons.org/dpubs_series/7777