CPC G06T 15/405 (2013.01) [G06T 19/006 (2013.01); G06T 2210/12 (2013.01)] | 20 Claims |
1. A method comprising, by a computing system:
accessing a scene graph defining a spatial relationship of a plurality of first bounding boxes associated with a plurality of virtual objects in a field of view of a user of an artificial reality (AR) device;
accessing a second bounding box associated with a first bounding box of the plurality of first bounding boxes;
determining, based on the second bounding box, a subregion within a sensing region of a depth sensor;
obtaining first depth data within the subregion;
generating, based on the first depth data, a three-dimensional model for one or more real objects;
determining a visibility of a first virtual object in an AR environment associated with the first bounding box based on the three-dimensional model for the one or more real objects; and
generating an output image of the AR environment based on the determined visibility of the first virtual object.
|