Abstract: A system for preserving or establishing an ecosystem comprises an assembly coupled to a guidance component. An assembly comprises an enclosure and a substrate, and is configured to contain one or more organisms within the enclosure. The guidance component causes the assembly to be transported to a destination environment. The assembly is decoupled from the guidance component. The assembly detaches from the substrate, to integrate the one or more organisms into an ecosystem of the destination environment.

Abstract: An underwater positioning system comprises a plurality of underwater beacons. A beacon, in response to a signal sent by an underwater vehicle, responds with a signal comprising one or more characteristics to identify the beacon. Components of an access algorithm are provided to the underwater vehicle. The access algorithm determines a location of the beacon based on the beacon's identity. A position of the vehicle is determined based at least in part on the location of the beacon.

Abstract: Systems and techniques for enhancing the physical games with augmented reality is described. A system identifies a characteristic of a physical object that is located on a physical surface. The system obtains, based on the characteristic, information indicative of a rule set for the game and a state of the game based on the characteristic. A game engine is initialized based on the rule set and game state. The game engine generates a subsequent state for the game. The system renders, based on the subsequent state, an augmented reality display that incorporates the physical surface and physical object.

Abstract: Methods and systems for intraoperatively determining alignment parameters of a spine during a spinal surgical procedure are disclosed herein. In some embodiments, a method of intraoperatively determining an alignment parameter of a spine during a surgical procedure includes receiving initial image data of the spine including multiple vertebrae and identifying a geometric feature associated with each vertebra in the initial image data. The geometric features each have a pose in the initial image data and characterize a three-dimensional (3D) shape of the associated vertebra. The method further comprises receiving intraoperative image data of the spine and registering the initial image data to the intraoperative image data. The method can then update the pose of each geometric feature based on the registration and the intraoperative image data, and determine the alignment parameter based on the updated poses of the geometric features associated with two or more of the vertebrae.

Abstract: An augmented-reality system classifies a subject observed in video data obtained from a first area. A goal of a user of an augmented reality device is determined based on video data obtained from a second area. A correlation between the goal of the user and the classification of the subject is determined. An augmented reality display is generated to include a visual indicia of the subject, the visual indicia generated to represent the correlation between the goal and the classification.

Abstract: A system for recognizing gestures generates a skeletal model from video data of a subject. A defined subset of attributes of the skeletal model are mapped to defined positions of a vector. A gesture is recognized by evaluating a neural network using the vector as input. The neural network, trained using training vectors generated according to the definitions of skeletal model attributes and vector positions, classifies a gesture based on the input vector.

Abstract: A first augmented reality device shares, with a second augmented reality device, a visualization of an object viewed by the respective devices in an augmented reality scene. The first augmented reality device receives property information, according to which the second augmented reality device is displaying the visualization. The first augmented reality device displays the visualization according to the property information, adjusted to account for differences in perspective between the first and second augmented reality devices.

Abstract: A kinematic analysis system captures and records participant motion via a plurality of video cameras. A participant feature and participant pose are identified in a frame of video data. The feature and pose are correlated across a plurality of frames. A three-dimensional path of the participant is determined based on correlating the feature and pose across the plurality of frames. A potential impact is identified based on analysis of the participant's path.

Abstract: A system for providing a virtual reality experience obtains, from a first sensor, image data of a first subject upon entry to an area. The system obtains pose information for the first subject during a period in which the subject is within the area. A graphical asset is generated from the image data. A skeletal model is maintained based on the pose information. A virtual reality scene is generated, using the asset and skeletal model, from the perspective of a second subject.

Abstract: An augmented reality system obtains sensor data depicting a first object in a scene. The identity of the first object is determined. The system estimates a size of the first object based on the determined identity. Based on the estimated size, the system estimates a scale of other objects depicted in the obtained sensor data. An augmented reality display is generated based at least in part on the size of a second object determined using the estimated scale.

Abstract: An augmented reality system associates a plurality of services with an augmented reality application that is executing on an augmented reality device. The system receives sensor data obtained by the augmented reality device, and causes the plurality of services to process the data. The system generates an event based on the processing of the received data by the services, and sends the event to the augmented reality device. The augmented reality device generates an augmented reality display based at least in part on the event.

Abstract: An augmented reality device enables selection and interaction with multiple objects viewed in an augmented reality scene. A cursor is positioned proximate to a first object in response to a first gesture. A selected state of the first object is determined in response to a second gesture. A visual indication of the selected state of the first object is displayed by the augmented reality device. A second position for the cursor, proximate to a second object, is determined. A visual indication of the first object's selected state remains displayed after the cursor is moved to be proximate to the second object.

Abstract: A system for recognizing gestures generates a skeletal model from video data of a subject. A defined subset of attributes of the skeletal model are mapped to defined positions of a vector. A gesture is recognized by evaluating a neural network using the vector as input. The neural network, trained using training vectors generated according to the definitions of skeletal model attributes and vector positions, classifies a gesture based on the input vector.

Abstract: An underwater positioning system comprises a plurality of underwater beacons. A beacon, in response to a signal sent by an underwater vehicle, responds with a signal comprising one or more characteristics to identify the beacon. Components of an access algorithm are provided to the underwater vehicle. The access algorithm determines a location of the beacon based on the beacon's identity. A position of the vehicle is determined based at least in part on the location of the beacon.

Abstract: A system for preserving or establishing an ecosystem comprises an assembly coupled to a guidance component. An assembly comprises an enclosure and a substrate, and is configured to contain one or more organisms within the enclosure. The guidance component causes the assembly to be transported to a destination environment. The assembly is decoupled from the guidance component. The assembly detaches from the substrate, to integrate the one or more organisms into an ecosystem of the destination environment.

Abstract: A system for providing a virtual reality experience obtains, from a first sensor, image data of a first subject upon entry to an area. The system obtains pose information for the first subject during a period in which the subject is within the area. A graphical asset is generated from the image data. A skeletal model is maintained based on the pose information. A virtual reality scene is generated, using the asset and skeletal model, from the perspective of a second subject.

Abstract: An augmented-reality system classifies a subject observed in video data obtained from a first area. A goal of a user of an augmented reality device is determined based on video data obtained from a second area. A correlation between the goal of the user and the classification of the subject is determined. An augmented reality display is generated to include a visual indicia of the subject, the visual indicia generated to represent the correlation between the goal and the classification.

Abstract: A diagnostic model, adapted to a condition of a patient, is maintained. A monitoring device obtains visual information of a patient. The visual information is converted to a format suitable as input to the diagnostic model. Output of the diagnostic model is used to generate a recommendation for at least one of diagnosis or treatment of the patient. A response to the diagnosis or treatment recommendation is generated.

Abstract: A visual characteristic on a physical surface in the physical space is determined from an image of a physical space, with the physical surface including a first area inside the visual characteristic and a second area outside of the visual characteristic. A first region of virtual space that corresponds to the first area and a second region of virtual space that corresponds to the second area is determined based at least in part on the visual characteristic. Information that includes a position and orientation of a digital object relative to the first region and the second region in the virtual space is determined based at least in part on the visual characteristic. The digital object is rendered at a position and orientation in the virtual space in accordance with the information.

Abstract: Systems and techniques for enhancing the physical games with augmented reality is described. A system identifies a characteristic of a physical object that is located on a physical surface. The system obtains, based on the characteristic, information indicative of a rule set for the game and a state of the game based on the characteristic. A game engine is initialized based on the rule set and game state. The game engine generates a subsequent state for the game. The system renders, based on the subsequent state, an augmented reality display that incorporates the physical surface and physical object.