Video encoding control using non-exclusive content categories
View Patent ↗In some embodiments, macroblock-level encoding parameters are assigned to weighted linear combinations of corresponding content-category-level encoding parameters. For example, a macroblock quantization parameter (QP) modulation is set to a weighted linear combination of content category QP modulations. Content categories may identify potentially overlapping content types such as sky, water, grass, skin, and red content. The combination weights may be similarity measures describing macroblock similarities to content categories. A macroblock may be associated with multiple content categories, with different similarity levels for different content categories. A similarity measure for a given macroblock with respect to a content category may be defined as a number (between 0 and 8) of neighboring macroblocks that meet a similarity condition, provided the macroblock meets a qualification condition. The similarity condition may be computationally simpler than the qualification condition. Content-category-level statistics are generated by combining block-level statistics. Content-category-level statistics may be used in encoding subsequent frames.
1. A video encoding method comprising:
setting a value of a block-level encoding parameter for a block in a video frame to a combination of a plurality of content-category-level extensions of the block-level encoding parameter; wherein the combination is a weighted linear combination according to a relation
p
m
=
K
∑
r
∈
R
W
m
,
r
P
r
wherein Pm is the block-level encoding parameter, an index m denotes the block, Pr denotes the plurality of content-category-level extensions of the block-level encoding parameter, Wm,r denotes a weight characterizing a similarity of the block m to a content category r, K is a constant, and R is a set of content categories r; and
encoding the block according to the value of the block-level encoding parameter.
2. The method of claim 1 , wherein the block-level encoding parameter comprises a block-level quantization precision indicator, and the plurality of content-category-level extensions of the block-level encoding parameter comprise a plurality of content-category level quantization precision indicators.
3. The method of claim 1 , wherein the block-level encoding parameter comprises a block-level quantization parameter modulation, and the plurality of content-category-level extensions of the block-level encoding parameter comprise a plurality of content-category level quantization parameter modulations.
4. A video encoding method comprising:
determining a plurality of encoding parameters for a corresponding plurality of content categories;
setting an encoding parameter for a block in a video frame to a combination of the plurality of encoding parameters defined for the corresponding plurality of content categories; wherein the combination is a weighted linear combination according to a relation
p
m
=
K
∑
r
∈
R
W
m
,
r
P
r
wherein Pm is the block-level encoding parameter, an index m denotes the block, Pr denotes the plurality of content-category-level extensions of the block-level encoding parameter, Wm,r denotes a weight characterizing a similarity of the block m to a content category r, K is a constant, and R is a set of content categories r; and
encoding the block according to the encoding parameter for the block.
5. The method of claim 4 , wherein each of the plurality of encoding parameters defined for the corresponding plurality of content categories comprises a content-category quantization precision indicator.
6. The method of claim 4 , wherein the combination is a linear function.
7. The method of claim 6 , wherein the linear function comprises a plurality of weights, each weight characterizing a similarity of the block to a content category.
8. The method of claim 7 , wherein said each weight characterizes a similarity of a set of neighbors of the block to the content category.
9. The method of claim 7 , wherein said each weight is proportional to a number of neighbors of the block that meet a similarity condition with respect to the content category.
10. The method of claim 4 , wherein the combination is a function comprising a plurality of weights, each weight characterizing a similarity of the block to a content category.
11. The method of claim 10 , wherein setting the encoding parameter comprises:
determining whether a content of the block meets a qualification condition with respect to the content category, wherein the qualification condition depends on a content of the block; and
provided the qualification condition is met, setting said each weight to be proportional to a number of neighbors of the block meeting a similarity condition with respect to the content category.
12. The method of claim 4 , wherein determining the plurality of encoding parameters comprises performing a rate-distortion analysis in a ρ-domain.
13. The method of claim 12 wherein each of the plurality of encoding parameters is a content category quantization parameter determined by interpolation from a set of quantization parameter-ρ correspondence values.
14. A video encoder comprising:
a content category processing unit setting a value of a block-level encoding parameter for a block in a video frame to a combination of a plurality of content-category-level extensions of the block-level encoding parameter; wherein the combination is a weighted linear combination according to a relation
p
m
=
K
∑
r
∈
R
W
m
,
r
P
r
wherein Pm is the block-level encoding parameter, an index m denotes the block, Pr denotes the plurality of content-category-level extensions of the block-level encoding parameter, Wm,r denotes a weight characterizing a similarity of the block m to a content category r, K is a constant, and R is a set of content categories r; and
encoding logic connected to the content category processing unit, the encoding logic encoding the block according to the value of the block-level encoding parameter.
15. A video encoder comprising:
means for setting a value of a block-level encoding parameter for a block in a video frame to a combination of a plurality of content-category-level extensions of the block-level encoding parameter; wherein the combination is a weighted linear combination according to a relation
p
m
=
K
∑
r
∈
R
W
m
,
r
P
r
wherein Pm is the block-level encoding parameter, an index m denotes the block, Pr denotes the plurality of content-category-level extensions of the block-level encoding parameter, Wm,r denotes a weight characterizing a similarity of the block m to a content category r, K is a constant, and R is a set of content categories r; and
means for encoding the block according to the value of the block-level encoding parameter.