Abstract: Example methods and systems for adjusting sensor viewpoint to a virtual viewpoint are provided. An example method may involve receiving data from a first camera; receiving data from a second camera; transforming, from the first viewpoint to a virtual viewpoint within the device, frames in a first plurality of frames based on an offset from the first camera to the virtual viewpoint; determining, in a second plurality of frames, one or more features and a movement, relative to the second viewpoint, of the one or more features; and transforming, from the second viewpoint to the virtual viewpoint, the movement of the one or more features based on an offset from the second camera to the virtual viewpoint; adjusting the transformed frames of the virtual viewpoint by an amount that is proportional to the transformed movement; and providing for display the adjusted and transformed frames of the first plurality of frames.

Abstract: Methods and systems for map generation are described. A computing device may receive outputs from a plurality of sensors at a position of the device in an environment, which may include data corresponding to visual features of the environment at the first position. Based on correspondence in the outputs from the plurality of sensors, the computing device may generate a map of the environment comprising sparse mapping data, and the sparse mapping data comprises the data corresponding to the visual features. The device may receive additional outputs at other positions of the device in the environment and may modify the map based on the additional outputs. In addition, the device may modify the map based on receiving dense mapping information from sensors, which may include data corresponding to objects in the environment in a manner such that represents a structure of the object in the environment.

Abstract: Methods and systems for determining estimation of motion of a device are provided. An example method includes receiving data from an inertial measurement unit (IMU) of a device and receiving images from a camera of the device for a sliding time window. The method also includes determining an IMU estimation of motion of the device based on the data from the IMU, and a camera estimation of motion of the device based on feature tracking in the images. The method includes, based on the IMU estimation and the camera estimation having a difference more than a threshold amount, determining one or more of a position or a velocity of the device for the sliding time window, and determining an overall estimation of motion of the device as supported by the data from the IMU and the position or velocity of the device.

Abstract: Example methods and systems for synchronizing data received from multiple sensors of a device are provided. A method may be performed by a device having an application processor configured to function based on an operating system and a co-processor configured to receive data from sensors of the device. The method may comprise determining an interrupt by a sensor of the device, and providing, by the co-processor, a timestamp of the interrupt that is indicative of a time that the sensor has data for output. The method also comprises receiving the data for output from the sensor, associating the timestamp of the interrupt by the sensor with the received data, associating together data received from multiple sensors into data structures based on timestamps of the data, and providing the data structures to the application processor in sequence based on the timestamps of the data.

Abstract: This disclosure describes techniques for reducing or eliminating estimator inconsistency in vision-aided inertial navigation systems (VINS). For example, an observability-constrained VINS (OC-VINS) is described which enforce the unobservable directions of the system to prevent spurious information gain and reduce inconsistency.

Abstract: A motion classification system comprises an inertial measurement unit configured to sense motion of a user and to output one or more channels of inertial motion data corresponding to the sensed motion; and a processing unit configured to calculate a coefficient vector for each of the one or more channels based on a wavelet transformation of the respective inertial motion data, and to select one of a plurality of gaits as the user's gait based on the calculated coefficient vector of at least one of the one or more channels and on a plurality of templates, each template corresponding to one of the plurality of gaits.

Abstract: Systems and methods for constructing distance estimate models for personal navigation are provided. In one embodiment, a distance estimation system comprises: a gait information memory configured to store gait information about a gait mode; a biometric data memory configured to store a biometric profile for a user; a frequency module configured to identify a gait frequency; and a distance calculation module configured to calculate the distance traveled by the user by creating a distance estimate model based on the gait mode, the biometric profile, and the gait frequency, wherein the distance calculation module creates the distance estimate model by performing a regression analysis on movement information from at least one user.

Abstract: A motion classification system comprises an inertial measurement unit configured to sense motion of a user and to output one or more channels of inertial motion data corresponding to the sensed motion; and a processing unit configured to calculate a coefficient vector for each of the one or more channels based on a wavelet transformation of the respective inertial motion data, and to select one of a plurality of gaits as the user's gait based on the calculated coefficient vector of at least one of the one or more channels and on a plurality of templates, each template corresponding to one of the plurality of gaits.