Systems and methods for adaptively partitioning video blocks for video coding

1. A method of reconstructing video data, the method comprising:

receiving a first value indicating a maximum allowed partitioning depth, the first value being a sequence parameter set (SPS) syntax element;

receiving a flag in the video data indicating whether the first value is overridden;

determining the maximum allowed partitioning depth based on a second value when the flag indicates that the first value is overridden, wherein a first syntax level of the first value is higher than a second syntax level of the second value;

determining a partitioning for a coded video block based on the maximum allowed partitioning depth; and

reconstructing the video data based on the determined partitioning for the coded video block.

2. The method of claim 1 , wherein the maximum allowed partitioning depth corresponds to a maximum number of times a quad tree leaf node is partitioned according to a binary split.

3. A device for decoding video data, the device comprising:

one or more non-transitory computer-readable media having executable instructions embodied thereon; and

at least one processor coupled to the one or more non-transitory computer-readable media and configured to execute the executable instructions to:

receive a first value indicating a maximum allowed partitioning depth, the first value being a sequence parameter set (SPS) syntax element;

receive a flag in the video data indicating whether the first value is overridden;

determine the maximum allowed partitioning depth based on a second value when the flag indicates that the first value is overridden, wherein a syntax level of the first value is higher than a syntax level of the second value;

determine a partitioning for a coded video block based on the maximum allowed partitioning depth; and

reconstruct the video data based on the determined partitioning for the coded video block.

4. The device of claim 3 , wherein the maximum allowed partitioning depth corresponds to a maximum number of times a quad tree node is partitioned according to a binary split.

5. A system for reconstructing video data, the system comprising:

a decoder comprising one or more non-transitory computer-readable media having executable instructions embodied thereon; and

at least one processor coupled to the one or more non-transitory computer-readable media and configured to execute the executable instructions to:

receive a first value indicating a maximum allowed partitioning depth, the first value being a sequence parameter set (SPS) syntax element;

receive a flag in the video data indicating whether the first value is overridden;

determine the maximum allowed partitioning depth based on a second value when the flag indicates that the first value is overridden, wherein a syntax level of the first value is higher than a syntax level of the second value;

determine a partitioning for a coded video block based on the maximum allowed partitioning depth; and

reconstruct the video data based on the determined partitioning for the coded video block.

6. The system of claim 5 , further comprising:

an encoder comprising one or more non-transitory computer-readable media having executable instructions embodied thereon; and

at least one processor coupled to the one or more non-transitory computer-readable media and configured to execute the executable instructions to:

receive a video block including sample values;

set the first value indicating the maximum allowed partitioning depth;

determine the maximum allowed partitioning depth based on the second value and signal the flag into the video data when the first value is overridden; and

partition the sample values according to the maximum allowed partitioning depth.

7. The system of claim 5 , wherein the maximum allowed partitioning depth corresponds to a maximum number of times a quad tree leaf node is partitioned according to a binary split.

8. The system of claim 6 , wherein the at least one processor of the encoder is further configured to execute the executable instructions to signal the first value and the second value.

9. The system of claim 6 , wherein partitioning the sample values comprises allowing or disallowing one or more subsequent partitioning modes based on the maximum allowed partitioning depth.

10. The method of claim 1 , further comprising determining the maximum allowed partitioning depth further based on one or more inference rules when the first value is overridden.

11. The method of claim 10 , wherein one of the one or more inference rules is defined for a combination of a component channel and a slice type.

12. The method of claim 11 , wherein:

the component channel is one of a luma channel and a chroma channel of the video data; and

the slice type is one of an intra slice and an inter slice.

13. The method of claim 1 , wherein:

the first value indicates the maximum allowed partitioning depth for a specific partitioning scheme; and

the second value overrides the first value for the specific partitioning scheme when the flag indicates that the first value is overridden.

14. The method of claim 13 , wherein the flag indicates that the maximum allowed partitioning depth for the specific partitioning scheme is inherited from the first value in a first syntax structure having the first syntax level or determined based on the second value in a second syntax structure having the second syntax level lower than the first syntax level.

15. The method of claim 1 , further comprising:

determining that the maximum allowed partitioning depth is equal to the first value without reference to the second value when the flag indicates that the first value is not overridden.

16. The method of claim 1 , wherein determining the maximum allowed partitioning depth based on the second value comprises:

determining the maximum allowed partitioning depth based on the second value without reference to the first value when the flag indicates that the first value is overridden.