Search for: All records

Sleep is a vital physiological state that significantly impacts overall health. Continuous monitoring of sleep posture, heart rate, respiratory rate, and body movement is crucial for diagnosing and managing sleep disorders. Current monitoring solutions often disrupt natural sleep due to discomfort or raise privacy and instrumentation concerns. We introduce PillowSense, a fabric-based sleep monitoring system seamlessly integrated into a pillowcase. PillowSense utilizes a dual-layer fabric design. The top layer comprises conductive fabrics for sensing electrocardiogram (ECG) and surface electromyogram (sEMG), while the bottom layer features pressure-sensitive fabrics to monitor sleep location and movement. The system processes ECG and sEMG signals sequentially to infer multiple sleep variables and incorporates an adversarial neural network to enhance posture classification accuracy. We fabricate prototypes using off-the-shelf hardware and conduct both lab-based and in-the-wild longitudinal user studies to evaluate the system's effectiveness. Across 151 nights and 912.2 hours of real-world sleep data, the system achieves an F1 score of 88% for classifying seven sleep postures, and clinically-acceptable accuracy in vital sign monitoring. PillowSense's comfort, washability, and robustness in multi-user scenarios underscore its potential for unobtrusive, large-scale sleep monitoring. 

Visual tags (e.g., barcodes, QR codes) are ubiquitous in modern day life, though they rely on obtrusive geometric patterns to encode data, degrading the overall user experience. We propose a new paradigm of passive visual tags which utilizes light polarization to imperceptibly encode data using cheap, widely-available components. The tag and its data can be extracted from background scenery using off-the-shelf cameras with inexpensive LCD shutters attached atop camera lenses. We examine the feasibility of this design with real-world experiments. Initial results show zero bit errors at distances up to 3.0~m, an angular-detection range of \ang110, and robustness to manifold ambient light and occlusion scenarios. 

This paper presents a holistic system to scale up the teaching and learning of vocabulary words of American Sign Language (ASL). The system leverages the most recent mixed-reality technology to allow the user to perceive her own hands in an immersive learning environment with first- and third-person views for motion demonstration and practice. Precise motion sensing is used to record and evaluate motion, providing real-time feedback tailored to the specific learner. As part of this evaluation, learner motions are matched to features derived from the Hamburg Notation System (HNS) developed by sign-language linguists. We develop a prototype to evaluate the efficacy of mixed-reality-based interactive motion teaching. Results with 60 participants show a statistically significant improvement in learning ASL signs when using our system, in comparison to traditional desktop-based, non-interactive learning. We expect this approach to ultimately allow teaching and guided practice of thousands of signs.