Abstract: A system and method for monitoring medication intake uses vibration signals generated by a kinetic augmentation attachment. The attachment includes a base coupled to a medication container, a spring, and a weight, which produces distinct vibration signatures when the container is placed on a surface. A vibration sensor detects these signals, and a computing device processes them to extract features of the attachment and identify the medication container. The system does not require batteries, enabling long-term, maintenance-free monitoring. The method includes processing vibration signals using frequency analysis and machine learning to recognize different medication containers based on their kinetic signatures.

Abstract: Cross-modal association (CMA), such as cross-modal signal segment association, is described for associating structural vibration and wearable sensors. This includes an Association Discovery Temporal Convolutional Network (AD-TCN) that determines the amount of shared context between a structural vibration sensor and associated wearable sensor candidates from the parameters of the trained model. CMA achieves improvements in AUC values, F1 scores, and accuracy over relevant baselines. In at least one embodiment, the vibration sensor is associated with a physical structure, the wearable sensor is associated with a person, and the modules estimate association between vibration sensor signals and a person who induces the vibration sensor signals interior to the structure.

Abstract: A smart retail product shelf liner called ‘MOOCA’ to detect product loads on its surface. The liner preferably has an origami like structure whose properties (e.g., angles, unit size, material) are adjusted to make the liner configurable for different sizes, load ranges, and product packaging. The smart liner is well suited for autonomous checkout systems. MOOCA is an origami-inspired low-cost configurable surface structure having conductive threads and copper wires integrated into the origami structure to detect and recognize product loads and thus to detect when products are either picked-up, or put-down, upon the smart MOOCA liner of the shelf.

Abstract: This invention introduces an indoor person identification system that utilizes the capture and analysis of footstep induced structural vibrations. The system senses floor vibration and detects the signal induced by footsteps. Then the system then extracts features from the signal that represent characteristics of each person's unique gait pattern. With these extracted features, the system conducts hierarchical classification at an individual step level and at a collection of consecutive steps level, achieving high degree of accuracy in the identification of individuals.

Abstract: A method and apparatus are described including receiving sensor data from at least one of a plurality of sensors in a sensor array associated with an environment, detecting a pattern in the received sensor data, determining a profile of at least one user present in the environment from a correlation between the detected pattern and patterns in a first database, determining an action to be performed responsive to the determined age range of the at least one user in the environment to actions associated with the user profile in a second database and performing the determined action on received content.

Abstract: Disclosed are systems, apparatus, devices, methods, computer program media and products, and other implementations, including a method that includes capturing an RGB image of a hand of a person, capturing a near infrared (NIR) image of the hand, and determining geometrical features of the hand based on the captured RGB and NIR images. In some embodiments, determining the geometrical features may include determining from extracted image data for the hand one or more values representative of relative spatial features for the hand such as one or more of, for example an angle between a first pair of two fingers in the extracted image data for the hand, and/or a relative distance or length for a second pair of fingers in the extracted image data for the hand. In some embodiments, the method may further include determining identity of the person based on the determined geometrical features of the hand.

Abstract: This invention introduces an indoor person identification system that utilizes the capture and analysis of footstep induced structural vibrations. The system senses floor vibration and detects the signal induced by footsteps. Then the system then extracts features from the signal that represent characteristics of each person's unique gait pattern. With these extracted features, the system conducts hierarchical classification at an individual step level and at a collection of consecutive steps level, achieving high degree of accuracy in the identification of individuals.

Abstract: Disclosed are systems, apparatus, devices, methods, computer program media and products, and other implementations, including a method that includes capturing an RGB image of a hand of a person, capturing a near infrared (NIR) image of the hand, and determining geometrical features of the hand based on the captured RGB and NIR images. In some embodiments, determining the geometrical features may include determining from extracted image data for the hand one or more values representative of relative spatial features for the hand such as one or more of, for example an angle between a first pair of two fingers in the extracted image data for the hand, and/or a relative distance or length for a second pair of fingers in the extracted image data for the hand. In some embodiments, the method may further include determining identity of the person based on the determined geometrical features of the hand.