Boundary detection using vision-based feature mapping

    公开(公告)号:US10970929B2

    公开(公告)日:2021-04-06

    申请号:US16512514

    申请日:2019-07-16

    申请人: Occipital, Inc.

    IPC分类号: G06T15/00 G06T19/00

    摘要: A system configured to determine one or more boundaries of the space, for example, in 3D applications. In some cases, the system may generate point data including a set of safe points in a space and a set of unsafe points in the space, the space surrounding a user device of the system, generate a triangulation over a union of the set of safe points and the set of unsafe points, determine triangles of the triangulation that include at least one safe point and determine edges of determined triangles which are part of a single triangle that include at least one safe point. The system may then determine one or more boundaries of the space using the determined edges.

    Mixed reality controller and headset tracking system

    公开(公告)号:US10679378B1

    公开(公告)日:2020-06-09

    申请号:US16254824

    申请日:2019-01-23

    申请人: Occipital, Inc.

    摘要: A system configured to determine a six-degree of freedom pose of a physical object in a physical environment and to utilize the six-degree of freedom pose as an within a virtual environment or mixed reality environment. In some cases, the system may utilize one or more cameras on a headset device to track the pose of a controller or other objects and one or more cameras on the controller itself to track the pose of the headset device or the user. In one example, the system may capture image data of a physical object having a constellation or pattern on the external source. The system may analyze the image data to identify image points associated with the constellation or pattern and to determine the pose of the object based on a location of the points in the image.

    SYSTEM AND METHOD FOR RELOCALIZATION AND SCENE RECOGNITION

    公开(公告)号:US20200050904A1

    公开(公告)日:2020-02-13

    申请号:US16655388

    申请日:2019-10-17

    申请人: Occipital, Inc.

    摘要: A system configured to improve the operations associated with generating virtual representations of physical environments to recognize the physical environments and/or relocalize within the virtual representations in a substantially real time system. In some cases, the system may use a first pre-training phase of descriptors and/or split nodes of regression forests using features common across various scenes to learn general image appearance, and a second training phase of descriptors and/or leaf nodes of regression forests to learn scene specific features. The system may align the features using an orientation vector, correct for camera perspective and lens distortion of the features as well as learn robust illumination invariant features from real and synthetic data.