| CPC G06T 7/543 (2017.01) [G06T 7/60 (2013.01); G06T 7/73 (2017.01); G06T 7/90 (2017.01); G06T 19/20 (2013.01); G06T 2207/10028 (2013.01)] | 39 Claims |
|
1. A method executed by one or more computing devices, the method comprising:
generating a view of a scene based on one or more images corresponding to one or more sets of positions in the scene;
identifying one or more objects included in the view;
generating geometric surface information for the view, the geometric surface information identifying one or more planar surfaces in the view;
generating a depthmap that includes depth estimates for each pixel of the view, wherein generating the depthmap comprises:
identifying a plurality of edge pixels included in edges of the one or more objects included in the view;
enhancing a depth of at least one edge pixel based at least in part on one or more depths of one or more other edge pixels included in the same edge;
identifying a plurality of planar surface pixels that are included in a planar surface in the one or more planar surfaces; and
for at least one planar surface pixel, adjusting a depth of at least one planar surface pixel in the plurality of planar surface pixels based at least in part on one or more depths of one or more other planar surface pixels in the plurality of planar surface pixels.
|