Abstract: Techniques and systems are provided for detecting object surfaces in extended reality environments. In some examples, a system obtains image data associated with a portion of a scene within a field of view (FOV) of a device. The portion of the scene includes at least one object. The system determines, based on the image data, a depth map of the portion of the scene. The system also determines, using the depth map, one or more planes within the portion of the scene. The system then generates, using the one or more planes, at least one planar region with boundaries corresponding to boundaries of a surface of the at least one object. The system also generates a three-dimensional representation of the portion of the scene using the at least one planar region and updates a three-dimensional representation of the scene using the three-dimensional representation of the portion of the scene.

Abstract: A visual tracker may track an object by identifying the object in a frame, and the visual tracker by identify the object in the frame within a search region. The search region may be provided by a motion modeling system that independently models the motion of the object and models the motion of the camera. For example, an object motion model of the motion modeling system may first model the motion of the object, assuming the camera is not in motion, in order to identify the expected position of the object. A camera motion model of the motion modeling system may then update the expected position of the object, obtained from the object motion model, based on the motion of the camera.

Abstract: A visual tracker may track an object by identifying the object in a frame, and the visual tracker by identify the object in the frame within a search region. The search region may be provided by a motion modeling system that independently models the motion of the object and models the motion of the camera. For example, an object motion model of the motion modeling system may first model the motion of the object, assuming the camera is not in motion, in order to identify the expected position of the object. A camera motion model of the motion modeling system may then update the expected position of the object, obtained from the object motion model, based on the motion of the camera.