Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for handing off a user conversation between computer-implemented agents. One of the methods includes receiving, by a computer-implemented agent specific to a user device, a digital representation of speech encoding an utterance, determining, by the computer-implemented agent, that the utterance specifies a requirement to establish a communication with another computer-implemented agent, and establishing, by the computer-implemented agent, a communication between the other computer-implemented agent and the user device.

Abstract: A system for predicting and summarizing medical events from electronic health records includes a computer memory storing aggregated electronic health records from a multitude of patients of diverse age, health conditions, and demographics including medications, laboratory values, diagnoses, vital signs, and medical notes. The aggregated electronic health records are converted into a single standardized data structure format and ordered arrangement per patient, e.g., into a chronological order. A computer (or computer system) executes one or more deep learning models trained on the aggregated health records to predict one or more future clinical events and summarize pertinent past medical events related to the predicted events on an input electronic health record of a patient having the standardized data structure format and ordered into a chronological order.

Abstract: This disclosure relates to analysing rock falls. A video camera captures video data, a depth measurement system captures depth data, a seismic sensor captures seismic data, and a data store stores spatial characterising data of the rock fall site. A processor detects a rock fall source location based on the video data, depth data, seismic data; determines a three-dimensional free fall estimation based on the rock fall source location; estimates three-dimensional bounce kinematics based on the three-dimensional free fall estimation and the spatial characterising data of the rock fall site; estimates a runout based on the rock fall source location, free fall and bounce kinematics; and combines the rock fall source location, the three-dimensional free fall, the three dimensional bounce kinematics and the runout to determine a rock fall trajectory.

Abstract: A system for moving an object within an environment, wherein the system includes at least one modular wheel configured to move the object. The modular wheel includes a body configured to be attached to the object, a wheel, a drive configured to rotate the wheel and a controller configured to control the drive. One or more processing devices configured are provided to receive an image stream including a plurality of captured images from each of a plurality of imaging devices, the plurality of imaging devices being configured to capture images of the object within the environment, analyse the images to determine an object location within the environment, generate control instructions at least in part using the determined object location and provide the control instructions to the controller, the controller being responsive to the control instructions to control the drive and thereby move the object.

Abstract: An organizer includes a generally planar back member, a hanging assembly, and a retention assembly. The hanging assembly is coupled to said back member and is structured to be further coupled to a vertical support. The retention assembly includes a number of retention elements wherein each retention element structured to maintain an object adjacent to said back member. Further, the number of retention elements includes a suspended stapler holder. The suspended stapler holder is structured to suspend a stapler with minimal contact.

Abstract: Systems and methods for performing MOS skin deformations are provided. In one example process, the in vector of a MOS transform may be manually configured by a user. In another example process, a slide/bulge operation may be configured to depend on two or more MOS transforms. Each of the MOS transforms may be assigned a weight that represents the transform's contribution to the overall slide/bulge. In yet another example process, a bulge operation for a MOS vertex may be performed in a direction orthogonal to the attached MOS curve regardless of the direction of the attachment vector. In yet another example process, a ghost transform may be inserted into a MOS closed curve and used to calculate skin deformations associated with first transform of the MOS closed curve.

Abstract: Systems and methods for performing MOS skin deformations are provided. In one example process, the in vector of a MOS transform may be manually configured by a user. In another example process, a slide/bulge operation may be configured to depend on two or more MOS transforms. Each of the MOS transforms may be assigned a weight that represents the transform's contribution to the overall slide/bulge. In yet another example process, a bulge operation for a MOS vertex may be performed in a direction orthogonal to the attached MOS curve regardless of the direction of the attachment vector. In yet another example process, a ghost transform may be inserted into a MOS closed curve and used to calculate skin deformations associated with first transform of the MOS closed curve.

Abstract: Motion estimation systems and methods are disclosed. An apparatus may include a processing unit to acquire video images and to arrange the video images into a plurality of sequential video frames, and a motion estimation unit that receives the sequential video frames and determines a set of repetitive pattern neighbor candidate vectors for repetitive pattern content in a first frame. The set of repetitive pattern neighbor candidate vectors may be reduced by sorting the set to eliminate spurious repetitive pattern neighbor candidate vectors. The reduced set may be provided to a second adjacent frame. A method may include acquiring a plurality of sequential video frames having a repetitive pattern content, and determining a set of repetitive pattern neighbor candidate vectors for the repetitive pattern content in a first frame of the sequential video frames. The set of repetitive pattern neighbor candidate vectors may be sorted by determining at least one spurious repetitive pattern neighbor candidate vector.

Abstract: Motion estimation systems and methods are disclosed. An apparatus may include a processing unit to acquire video images and to arrange the video images into a plurality of sequential video frames, and a motion estimation unit that receives the sequential video frames and determines a set of repetitive pattern neighbor candidate vectors for repetitive pattern content in a first frame. The set of repetitive pattern neighbor candidate vectors may be reduced by sorting the set to eliminate spurious repetitive pattern neighbor candidate vectors. The reduced set may be provided to a second adjacent frame. A method may include acquiring a plurality of sequential video frames having a repetitive pattern content, and determining a set of repetitive pattern neighbor candidate vectors for the repetitive pattern content in a first frame of the sequential video frames. The set of repetitive pattern neighbor candidate vectors may be sorted by determining at least one spurious repetitive pattern neighbor candidate vector.

Abstract: The present disclosure provides a system and method for detecting motion vectors in an image frame using a recursive hierarchical process with a non-rasterized vector-scanning motion to reduce erroneous motion vectors in an image frame of a digital video sequence. In general, a resolution hierarchy is generated for an image frame, wherein the resolution hierarchy comprises the original image frame and one or more copy image frames each having a different, lower resolution than the original image frame. Each image frame in the hierarchy is partitioned into image patches disposed in columns and rows, and the image patches are scanned in a non-rasterized motion to detect motion vectors in each image patch. The disclosed system and method provides faster convergence and improved accuracy by converging motion vectors in multiple directions and minimizing erroneous motion vectors in the image sequence.

Abstract: A toy glider including a shaft, a roller attached to a first end of the shaft, and a housing attached to a second opposing end of the shaft. The roller, a front end portion of the housing, and other portions of the toy glider are interchangeable to create a variety of different designs.

Abstract: A boat propeller comprising a central hub member and an inner hub assembly that defines a longitudinally extending bore having an inner surface. The exterior surface of the central hub member is sized and shaped for disposition therein the bore of the inner hub assembly in a complementary fashion. In one aspect, the propeller may also comprise a plurality of resilient spacer members positioned such that that the exterior surface of the central hub member is spaced from the inner surface of the bore.