Abstract: A method, system and computer product for training a control input system involve taking an integral of an output value from a Motion Decision Neural Network for one or more movable joints to generate an integrated output value and generating a subsequent output value using a machine learning algorithm that includes a sensor value and a previous joint position if the integrated output value does not at least meet the threshold. Surface damping interactions with at least a simulated environment, a rigid body position and a position of the one or more movable joints based on an integral of the subsequent output value are simulated. The Motion Decision Neural Network is trained with the machine learning algorithm based upon at least a result of the simulation of the simulated environment and position of the one or more movable joints.

Abstract: A method, system and computer program product for training a control input system involve taking an integral of an output value from a Motion Decision Neural Network for one or more movable joints to generate an integrated output value and comparing the integrated output value to a backlash threshold. A subsequent output value is generated using a machine learning algorithm that includes a sensor value and a previous joint position if the integrated output value does not at least meet the threshold. A position of the one or more movable joints is simulated based on an integral of the subsequent output value; and the Motion Decision Neural Network is trained with the machine learning algorithm based upon at least a result of the simulation of the position of the one or more movable joints.

Abstract: Computer animation involving pose disambiguation is disclosed. Two or more source segmentation masks are generated from corresponding contemporaneous video images of a character from different points of view. A three-dimensional model of an animation character corresponding to the character in the two or more contemporaneous video images is generated. Two or more different target segmentation masks corresponding to different views of the animation character corresponding to the character in the two or more video images. Each target segmentation mask is compared to a corresponding source segmentation mask from the comparison it is determined whether a pose of the three-dimensional model corresponds to a pose of the character in the video images. The model is used to generate a frame of animation of the animated character when the pose of model corresponds to the pose of the character in the video images.

Abstract: A method, system and computer product for training a control input system involve taking an integral of an output value from a Motion Decision Neural Network for one or more movable joints to generate an integrated output value and generating a subsequent output value using a machine learning algorithm that includes a sensor value and a previous joint position if the integrated output value does not at least meet the threshold. Surface stiffness interactions with at least a simulated environment, a rigid body position and a position of the one or more movable joints based on an integral of the subsequent output value are simulated. The Motion Decision Neural Network is trained with the machine learning algorithm based upon at least a result of the simulation of the simulated environment and position of the one or more movable joints.

Abstract: Computer animation involving monocular pose prediction is disclosed. A plurality of candidate pose sequences of a three-dimensional model of an animation character is generated such that each candidate pose of each sequence has a segmentation map that matches a segmentation map of a corresponding character derived from a corresponding frame of a video. A distance between candidate poses at each time step is maximized. An optimum pose sequence is determined and used to generate a corresponding sequence of frames of animation.

Abstract: An event driven sensor (EDS) is used for simultaneous localization and mapping (SLAM) and in particular is used in conjunction with a constellation of light emitting diodes (LED) to simultaneously localize all LEDs and track EDS pose in space. The EDS may be stationary or moveable and can track moveable LED constellations as rigid bodies. Each individual LED is distinguished at a high rate using minimal computational resources (no image processing). Thus, instead of a camera and image processing, rapidly pulsing LEDs detected by the EDS are used for feature points such that EDS events are related to only one LED at a time.

Abstract: To derive three dimensional (3D) position and orientation of a 3-axis (or more) magnetometer/accelerometer device (such as may be implemented in VR or AR headset or computer game controller) without line of sight constraints, a spinning magnetic field is used to discriminate and remove the external (Earth's) magnetic field from the spinning magnetic field. This reduces the problem to finding the distance to the source of the magnetic field using a calibration table (or formula), finding two angles describing the deviation of the magnetic sensor from the axis of rotation of the spinning magnetic field and the phase around this axis, and from these values deriving the orientation of the sensor.

Abstract: Motion sensor assemblies are provided on the head and in both hands of a person to generate pose information. The pose information is used to enter a database of animations of whole skeleton bone poses that correlates signals from the three assemblies to whole body pose signals. The closest matching frame in the database and subsequent frames are used to provide a whole-body animation sequence based on the signals from the three motion sensor assemblies.

Abstract: Techniques for transferring highly dimensional movements to lower dimensional robot movements are described. In an example, a reference motion of a target is used to train a non-linear approximator of a robot to learn how to perform the motion. The robot and the target are associated with a robot model and a target model, respectively. Features related to the positions of the robot joints are input to the non-linear approximator. During the training, a robot joint is simulated, which results in movement of this joint and different directions of a robot link connected thereto. The robot link is mapped to a link of the target model. The directions of the robot link are compared to the direction of the target link to learn the best movement of the robot joint. The training is repeated for the different links and for different phases of the reference motion.

Abstract: An event driven sensor (EDS) is used for simultaneous localization and mapping (SLAM) and in particular is used in conjunction with a constellation of light emitting diodes (LED) to simultaneously localize all LEDs and track EDS pose in space. The EDS may be stationary or moveable and can track moveable LED constellations as rigid bodies. Each individual LED is distinguished at a high rate using minimal computational resources (no image processing). Thus, instead of a camera and image processing, rapidly pulsing LEDs detected by the EDS are used for feature points such that EDS events are related to only one LED at a time.

Abstract: A method, system and non-transitory instructions for control input, comprising, taking an integral of an output value from a Motion Decision Neural Network for a movable joint to generate an integrated output value. Generating a subsequent output value using a machine learning algorithm that includes a sensor value and the integrated output value as inputs to the Motion Decision Neural Network and imparting movement with the moveable joint according to an integral of the subsequent output value.

Abstract: Techniques for transferring highly dimensional movements to lower dimensional robot movements are described. In an example, a reference motion of a target is used to train a non-linear approximator of a robot to learn how to perform the motion. The robot and the target are associated with a robot model and a target model, respectively. Features related to the positions of the robot joints are input to the non-linear approximator. During the training, a robot joint is simulated, which results in movement of this joint and different directions of a robot link connected thereto. The robot link is mapped to a link of the target model. The directions of the robot link are compared to the direction of the target link to learn the best movement of the robot joint. The training is repeated for the different links and for different phases of the reference motion.

Abstract: Plural individual sensor assemblies are engaged with respective parts of a person's body. Each assembly may include accelerometers, magnetometers, and gyroscopes. Sensor data is fused together to get the orientation at each body location. To simplify, the body is assumed to consist of rigid bars of known length connected with ball joints so that once the relative orientations of all bars are given by the respective assemblies, body pose can be computed. Then the body pose is translated as a virtual body into a virtual world either by a ray cast method that anchors a foot of the virtual body to the ground assuming infinite gravity and infinite friction and then translating the other body parts to make the ground contact point fixed, or by implementing an approximate dynamics physics engine on the virtual body. The technique may be used in VR location-based entertainment and for motion capture.

Abstract: To derive three dimensional (3D) position and orientation of a 3-axis (or more) magnetometer/accelerometer device (such as may be implemented in VR or AR headset or computer game controller) without line of sight constraints, a spinning magnetic field is used to discriminate and remove the external (Earth's) magnetic field from the spinning magnetic field. This reduces the problem to finding the distance to the source of the magnetic field using a calibration table (or formula), finding two angles describing the deviation of the magnetic sensor from the axis of rotation of the spinning magnetic field and the phase around this axis, and from these values deriving the orientation of the sensor.

Abstract: To derive three dimensional (3D) position and orientation of a 3-axis (or more) magnetometer/accelerometer device (such as may be implemented in VR or AR headset or computer game controller) without line of sight constraints, a spinning magnetic field is used to discriminate and remove the external (Earth's) magnetic field from the spinning magnetic field. This reduces the problem to finding the distance to the source of the magnetic field using a calibration table (or formula), finding two angles describing the deviation of the magnetic sensor from the axis of rotation of the spinning magnetic field and the phase around this axis, and from these values deriving the orientation of the sensor.

Abstract: Plural individual sensor assemblies are engaged with respective parts of a person's body. Each assembly may include accelerometers, magnetometers, and gyroscopes. Sensor data is fused together to get the orientation at each body location. To simplify, the body is assumed to consist of rigid bars of known length connected with ball joints so that once the relative orientations of all bars are given by the respective assemblies, body pose can be computed. Then the body pose is translated as a virtual body into a virtual world either by a ray cast method that anchors a foot of the virtual body to the ground assuming infinite gravity and infinite friction and then translating the other body parts to make the ground contact point fixed, or by implementing an approximate dynamics physics engine on the virtual body. The technique may be used in VR location-based entertainment and for motion capture.

Abstract: To derive three dimensional (3D) position and orientation of a 3-axis (or more) magnetometer/accelerometer device (such as may be implemented in VR or AR headset or computer game controller) without line of sight constraints, a spinning magnetic field is used to discriminate and remove the external (Earth's) magnetic field from the spinning magnetic field. This reduces the problem to finding the distance to the source of the magnetic field using a calibration table (or formula), finding two angles describing the deviation of the magnetic sensor from the axis of rotation of the spinning magnetic field and the phase around this axis, and from these values deriving the orientation of the sensor.

Abstract: An autonomous personal companion utilizing a method of object identification that relies on a hierarchy of object classifiers for categorizing one or more objects in a scene. The classifier hierarchy is composed of a set of root classifiers trained to recognize objects based on separate generic classes. Each root acts as the parent of a tree of child nodes, where each child node contains a more specific variant of its parent object classifier. The method covers walking the tree in order to classify an object based on more and more specific object features. The system is further comprised of an algorithm designed to minimize the number of object comparisons while allowing the system to concurrently categorize multiple objects in a scene.

Abstract: An autonomous personal companion executing a method including capturing data related to user behavior. Patterns of user behavior are identified in the data and classified using predefined patterns associated with corresponding predefined tags to generate a collected set of one or more tags. The collected set is compared to sets of predefined tags of a plurality of scenarios, each to one or more predefined patterns of user behavior and a corresponding set of predefined tags. A weight is assigned to each of the sets of predefined tags, wherein each weight defines a corresponding match quality between the collected set of tags and a corresponding set of predefined tags. The sets of predefined tags are sorted by weight in descending order. A matched scenario is selected for the collected set of tags that is associated with a matched set of predefined tags having a corresponding weight having the highest match quality.

Abstract: To derive three dimensional (3D) position and orientation of a 3-axis (or more) magnetometer/accelerometer device (such as may be implemented in VR or AR headset or computer game controller) without line of sight constraints, a spinning magnetic field is used to discriminate and remove the external (Earth's) magnetic field from the spinning magnetic field. This reduces the problem to finding the distance to the source of the magnetic field using a calibration table (or formula), finding two angles describing the deviation of the magnetic sensor from the axis of rotation of the spinning magnetic field and the phase around this axis, and from these values deriving the orientation of the sensor.