Modifying depth maps

1 . An apparatus for modifying depth maps, the apparatus comprising:

at least one memory; and

at least one processor coupled to the at least one memory and configured to:

obtain a depth map comprising a plurality of depth values;

obtain a plurality of confidence values comprising a respective confidence value for each depth value of the plurality of depth values;

divide the depth map into a plurality of depth sections;

determine, based on the plurality of confidence values, a plurality of section confidences comprising a respective section confidence for each of the plurality of depth sections; and

modify the depth map based on the plurality of section confidences, wherein to modify the depth map, the at least one processor is configured to modify low-confidence depth sections, each of the low-confidence depth sections having a respective section confidence less than a section-confidence threshold, and skip modifying of high-confidence depth sections, each of the high-confidence depth sections having a respective section confidence greater than the section-confidence threshold.

2 . The apparatus of claim 1 , wherein, to modify the depth map, the at least one processor is configured to apply a two-dimensional filter to depth values of the depth map.

3 . The apparatus of claim 1 , wherein, to modify the depth map, the at least one processor is configured to apply a weighted-averaging filter to depth values of the depth map.

4 . The apparatus of claim 1 , wherein, to modify the depth map, the at least one processor is configured to apply a smoothing filter to depth values of the depth map.

5 . The apparatus of claim 1 , wherein, to modify the depth map, the at least one processor is configured to generate filling depth values for one or more depth holes in the depth map based on neighboring depth values.

6 . The apparatus of claim 5 , wherein:

the depth map comprises a two-dimensional array of depth values;

the one or more depth holes comprise respective points in the two-dimensional array that lack respective depth values;

the neighboring depth values comprise depth values in the two-dimensional array that are adjacent to one or more respective depth holes; and

to generating the filling depth values for the one or more depth holes in the depth map, the at least one processor is configured to determine a respective filling depth value for each depth hole based on respective neighboring depth values of each depth hole.

7 . The apparatus of claim 6 , wherein the filling depth values are generated by applying a two-dimensional filter to the neighboring depth values.

8 . The apparatus of claim 1 , wherein a section confidence of a depth section is determined based on a statistical measure of confidence values of depth values of the depth section.

9 . The apparatus of claim 8 , wherein the statistical measure is based on at least one of:

an average of the confidence values; or

a minimum of the confidence values.

10 . The apparatus of claim 1 , wherein the at least one processor is further configured to determine a size for each of the plurality of depth sections based on the plurality of confidence values.

11 . The apparatus of claim 1 , wherein the at least one processor is further configured to determine a size for each of the plurality of depth sections based on a count of low-confidence depth values, wherein the low-confidence depth values correspond to respective confidence values that are less than a confidence threshold.

12 . A method for modifying depth maps, the method comprising:

obtaining a depth map comprising a plurality of depth values;

obtaining a plurality of confidence values comprising a respective confidence value for each depth value of the plurality of depth values;

dividing the depth map into a plurality of depth sections;

determining, based on the plurality of confidence values, a plurality of section confidences comprising a respective section confidence for each of the plurality of depth sections; and

modifying the depth map based on the plurality of section confidences, wherein modifying the depth map comprises modifying low-confidence depth sections, each of the low-confidence depth sections having a respective section confidence less than a section-confidence threshold, and skipping modifying of high-confidence depth sections, each of the high-confidence depth sections having a respective section confidence greater than the section-confidence threshold.

13 . The method of claim 12 , wherein modifying the depth map comprises applying a smoothing filter to depth values of the depth map.

14 . The method of claim 12 , wherein modifying the depth map comprises generating filling depth values for one or more depth holes in the depth map based on neighboring depth values.

15 . The method of claim 14 , wherein:

the depth map comprises a two-dimensional array of depth values;

the one or more depth holes comprise respective points in the two-dimensional array that lack respective depth values;

the neighboring depth values comprise depth values in the two-dimensional array that are adjacent to one or more respective depth holes; and

generating the filling depth values for the one or more depth holes in the depth map comprises determining a respective filling depth value for each depth hole based on respective neighboring depth values of each depth hole.

16 . The method of claim 15 , wherein the filling depth values are generated by applying a two-dimensional filter to the neighboring depth values.

17 . The method of claim 12 , further comprising determining a size for each of the plurality of depth sections based on the plurality of confidence values.

18 . The method of claim 12 , further comprising determining a size for each of the plurality of depth sections based on a count of low-confidence depth values, wherein the low-confidence depth values correspond to respective confidence values that are less than a confidence threshold.