FaceOri: Tracking Head Position and Orientation Using Ultrasonic Ranging on Earphones

GazeReader: Detecting Unknown Word Using Webcam for English as a Second Language (ESL) Learners

“…The optimization of the eye-tracking algorithm is needed to make gaze contribute more to the performance. Multi-modal data obtained by earphones and smart glasses can be used to infer head orientation and position [28,29] to facilitate the eye-tracking algorithm.…”

Semantic Hearing: Programming Acoustic Scenes with Binaural Hearables

“…Recent work has used in-ear sensors for health applications [11][12][13] and activity tracking [39,52]. Prior work has also explored various interaction modalities like ultrasound sensing [68] and on-face interaction [70] for in-ear devices. The closest to our work is Clearbuds [14], which focuses on the task of enhancing the speech of the wearer using synchronized audio signals from two wireless earbuds.…”

Deep learning and transfer learning for device-free human activity recognition: A survey

“…Using Channel Impulse Response (CIR) for data representation, UltraGesture achieves a range resolution of 7mm and recognizes 12 fine-grained hand gestures with an average accuracy of over 99%. In addition to hand gesture recognition, ultrasonic-based HAR systems have also been built for a variety of other applications such as in-air digits recognition [64] and head tracking [65], exhibiting high potentials in HAR tasks featuring short-distance interaction and fine-grained movements. In the wearable domain, IMU is the dominant or even the sole sensor for motion sensing in current commercial products.…”