Abstract: Techniques for generating a machine learning model to detect event instances from physical sensor data, including applying a first machine learning model to first sensor data from a first physical sensor at a location to detect an event instance, determining that a performance metric for use of the first machine learning model is not within an expected parameter, obtaining second sensor data from a second physical sensor during a period of time at the same location as the first physical sensor, obtaining third sensor data from the first physical sensor during the period of time, generating location-specific training data by selecting portions of the third sensor data based on training event instances detected using the second sensor data, training a second ML model using the location-specific training data, and applying the second ML model instead of the first ML model for detecting event instances.

Abstract: A computer system is provided that includes a server configured to store a plurality of location accounts, each location account being associated with a physical space at a recorded geospatial location. The plurality of location accounts utilize shared data definitions of a physical space parameter. Each physical space is equipped with a corresponding on-premise sensor configured to detect measured values for the physical space parameter over time and send to the server a data stream indicating the measured values. The computer system further includes a network portal, via which an authorized user for a location account can selectively choose whether to share the measured values or a summary thereof with other location accounts via the network portal.

Abstract: A computing system is provided, including a first display device having a first hardware configuration including a first display and a second display device having a second hardware configuration different from the first hardware configuration and including a second display. The computing system may further include a processor configured to receive an input including instructions to launch an application program on the first display device. The application program may include application program hardware specifications indicating hardware used by the application program. Based on the first hardware configuration, the second hardware configuration, and the application program hardware specifications, the processor may be further configured to determine that the second hardware configuration matches the application program hardware specifications more closely than the first hardware configuration. The processor may be further configured to launch the application program on the second display device.

Abstract: A wearable device is configured with a one-dimensional depth sensor (e.g., a LIDAR system) that scans a physical environment, in which the wearable device and depth sensor generate a point cloud structure using scanned points of the physical environment to develop blueprints for a negative space of the environment. The negative space includes permanent structures (e.g., walls and floors), in which the blueprints distinguish permanent structures from temporary objects. The depth sensor is affixed in a static position on the wearable device and passively scans a room according to the gaze direction of the user. Over a period of days, weeks, months, or years the blueprint continues to supplement the point cloud structure and update points therein. Thus, as the user continues to navigate the physical environment, over time, the point cloud data structure develops an accurate blueprint of the environment.

Abstract: The present disclosure provides apparatus and methods for automated tracking and counting of objects in a set of image frames using a resource-constrained device based on analysis of a selected subset of image frames, and based on selectively timing when resource-intensive operations are performed.

Abstract: A computer system is provided that includes a camera device configured to capture images of a physical scene and a processor configured to store a list of identified blobs. Each identified blob includes a determined blob feature and a path of tracked positions for that identified blob. The processor is configured to receive a series of images of the physical scene from the camera device, detect a candidate blob in a candidate image of the series of images, determine a blob feature of the candidate blob and a position of the candidate blob in the candidate image captured by the camera device. For each identified blob in the list, the processor is configured to compute a match score between the candidate blob and that identified blob.

Abstract: A wearable device is configured with various sensory devices that recurrently monitor and gather data for a physical environment surrounding a user to help locate and track real-world objects. The various heterogeneous sensory devices digitize objects and the physical world. Each sensory device is configured with a threshold data change, in which, when the data picked up by one or more sensory devices surpasses the threshold, a query is performed on each sensor graph or sensory device. The queried sensor graph data is stored within a node in a spatial graph, in which nodes are connected to each other using edges to create spatial relationships between objects and spaces. Objects can be uploaded into an object graph associated with the spatial graph, in which the objects are digitized with each of the available sensors. This digital information can be subsequently used to, for example, locate the object.

Abstract: To address issues of capturing and processing images, a mobile computing device is provided. The mobile computing device may include a two-part housing coupled by a hinge, with first and second parts that include first and second displays, respectively. The hinge may permit the displays to rotate throughout a plurality of angular orientations. The mobile computing device may include one or more sensor devices, processor, first camera, and second camera mounted in the housing. The one or more sensor devices may be configured to measure the relative angular displacement of the housing, and the processor may be configured to process images captured by the first and second cameras according to a selected function based upon the relative angular displacement measured by the one or more sensor devices.

Abstract: A computer system is provided that includes a camera device configured to capture images of a physical scene and a processor configured to store a list of identified blobs. Each identified blob includes a determined blob feature and a path of tracked positions for that identified blob. The processor is configured to receive a series of images of the physical scene from the camera device, detect a candidate blob in a candidate image of the series of images, determine a blob feature of the candidate blob and a position of the candidate blob in the candidate image captured by the camera device. For each identified blob in the list, the processor is configured to compute a match score between the candidate blob and that identified blob.

Abstract: A computer system is provided that includes a server configured to store a plurality of location accounts, each location account being associated with a physical space at a recorded geospatial location. The plurality of location accounts utilize shared data definitions of a physical space parameter. Each physical space is equipped with a corresponding on-premise sensor configured to detect measured values for the physical space parameter over time and send to the server a data stream indicating the measured values. The server further includes a promotion engine configured to determine similarities between measured values of the physical space parameter for the plurality of location accounts and generate co-promotions between location accounts of the plurality of location accounts based on the determined similarities. The computer system further includes a user interface system, via which an authorized user for a location account can view and accept a co-promotion.

Abstract: Examples are disclosed that relate to a wearable device configured to authenticate a user, associate itself with the user, and then authenticate the user to one or more other systems. One example provides a wearable device including a communication subsystem, a logic subsystem, and a storage subsystem including instructions executable to receive an input of information identifying a user, based upon the input, authenticate the user as a known user within a physical environment associated with the wearable device, associate the user with the wearable device, conduct an authentication communication with another computing device within the physical environment via the communication subsystem to authenticate the user to the other computing device via the wearable device, dissociate the wearable device from the user upon occurrence of an end-of-use session event, and later authenticate a different user and associate the different user with the wearable device.

Abstract: A wearable computing device is provided, comprising a camera and a microphone operatively coupled to a processor. Using both camera image data and speech recognition data, an object is detected and classified as an inventory item and inventory event. The inventory item and inventory event are subsequently recorded into an inventory database. Classifiers used to determine the inventory item and inventory event from the image data and speech may be cross trained based on the relative confidence values associated with each.

Abstract: Examples are disclosed that relate to conducting inventory management via wearable devices. One example provides a wearable device comprising a communication subsystem, one or more sensors, a logic subsystem, and a storage subsystem comprising instructions executable by the logic subsystem to receive an input activating the wearable device, receive via a sensor of the one or more sensors an input of information regarding a mark-out to make to inventory, provide an output confirming that the input of information was sensed, and send the information regarding the mark-out to make to inventory to an external computing device.

Abstract: A wearable device is configured with a one-dimensional depth sensor (e.g., a LIDAR system) that scans a physical environment, in which the wearable device and depth sensor generate a point cloud structure using scanned points of the physical environment to develop blueprints for a negative space of the environment. The negative space includes permanent structures (e.g., walls and floors), in which the blueprints distinguish permanent structures from temporary objects. The depth sensor is affixed in a static position on the wearable device and passively scans a room according to the gaze direction of the user. Over a period of days, weeks, months, or years the blueprint continues to supplement the point cloud structure and update points therein. Thus, as the user continues to navigate the physical environment, over time, the point cloud data structure develops an accurate blueprint of the environment.

Abstract: A wearable device is configured with various sensory devices that recurrently monitor and gather data for a physical environment surrounding a user to help locate and track real-world objects. The various heterogeneous sensory devices digitize objects and the physical world. Each sensory device is configured with a threshold data change, in which, when the data picked up by one or more sensory devices surpasses the threshold, a query is performed on each sensor graph or sensory device. The queried sensor graph data is stored within a node in a spatial graph, in which nodes are connected to each other using edges to create spatial relationships between objects and spaces. Objects can be uploaded into an object graph associated with the spatial graph, in which the objects are digitized with each of the available sensors. This digital information can be subsequently used to, for example, locate the object.

Abstract: In many computing scenarios, multiple users share a display to view and/or interact with content. Typically, one user provides input that interacts with the content; a second user can interact with the view only if the first user cedes control. Some interfaces permit split views, but typically support only a single input device that manipulates both panes. In the present disclosure, when a second user desires a different view of content during the first user's interaction, the device inserts a second view of the content into the display. Each view is associated with and manipulated by a particular user (e.g., via input devices associated with individual users) without altering the views of other users. The device may automatically manage the concurrent views, such as positioning and resizing; reflecting each user's perspective in other users' views; merging content changes; and terminating a view due to idleness or merging with another view.

Abstract: A computer system is provided that includes a server configured to store a plurality of location accounts, each location account being associated with a physical space at a recorded geospatial location. The plurality of location accounts utilize shared data definitions of a physical space parameter. Each physical space is equipped with a corresponding on-premise sensor configured to detect measured values for the physical space parameter over time and send to the server a data stream indicating the measured values. The computer system further includes a network portal, via which an authorized user for a location account can selectively choose whether to share the measured values or a summary thereof with other location accounts via the network portal.

Abstract: Contemporary human-computer interactions include conversational interactions, wherein devices present conversational prompts (e.g., generated speech) and conversational responses to user inquiries (e.g., verbal user input). Presented herein are techniques for automatically assembling conversational representations of web content. A variety of automated assembly techniques are disclosed, such as conversational template for websites of various website types. Interactions of users with a website may be monitored to identify actions that the users frequently perform, and conversational interactions may be generated that correspond to the actions. A web service may present a set of requests, and conversational interactions may be assembled to match the respective requests and responses of the web service. Conversational interactions may include transitions between websites, and conversational representations may be merged to integrate content from multiple websites.

Abstract: Techniques for generating a machine learning model to detect event instances from physical sensor data, including applying a first machine learning model to first sensor data from a first physical sensor at a location to detect an event instance, determining that a performance metric for use of the first machine learning model is not within an expected parameter, obtaining second sensor data from a second physical sensor during a period of time at the same location as the first physical sensor, obtaining third sensor data from the first physical sensor during the period of time, generating location-specific training data by selecting portions of the third sensor data based on training event instances detected using the second sensor data, training a second ML model using the location-specific training data, and applying the second ML model instead of the first ML model for detecting event instances.

Abstract: A computing device is provided that includes a primary display and a secondary display operatively coupled to a processor. The processor may be configured to execute an application program that has a GUI with a single display mode and a selectively displayable multiple display mode. The multiple display mode may include at least a primary view and a secondary view. In a single display mode, the processor may be configured to initially display the GUI on the primary display and not display the GUI on the secondary display. Upon receiving a multidisplay command to display the GUI in the multiple display mode, the processor may transition the application to the multiple display mode in which the primary view is displayed on the primary display, and the secondary view is displayed on the secondary display.