Efficient context model computation design in transform coefficient coding
A processor is configured to maintain, for encoding current values related to the transform coefficients a first line buffer and a second line buffer. The current values are arranged along a current scan-order anti-diagonal line. The first line buffer includes first values of a first scan-order anti-diagonal line. The second line buffer includes second values of a second scan-order anti-diagonal line. The processor is further configured to interleave the first values and the second values in a destination buffer; select, using the destination buffer, a probability distribution for coding a current value of the current values; entropy encode, in a compressed bitstream, the current value using the probability distribution; and replace, for coding values of an immediately subsequent scan-order anti-diagonal line to the current scan-order anti-diagonal line, one of the second line buffer or the first line buffer with the current scan-order anti-diagonal line.
1. An apparatus for encoding a transform block of transform coefficients, the apparatus comprising:
a processor configured to:
maintain, for encoding current values related to the transform coefficients a first line buffer and a second line buffer,
wherein the current values are arranged along a current scan-order anti-diagonal line of a scan order,
wherein the first line buffer includes first values of a first scan-order anti-diagonal line scanned immediately before the current scan-order anti-diagonal line, and
wherein the second line buffer includes second values of a second scan-order anti-diagonal line scanned immediately before the first scan-order anti-diagonal line;
interleave the first values of the first line buffer and the second values of the second line buffer in a destination buffer;
select, using the destination buffer, a probability distribution for coding a current value of the current values;
entropy encode, in a compressed bitstream, the current value using the probability distribution; and
replace, for coding values of an immediately subsequent scan-order anti-diagonal line to the current scan-order anti-diagonal line, one of the second line buffer or the first line buffer with the current scan-order anti-diagonal line.
2. The apparatus of claim 1 , wherein to interleave the first values of the first line buffer and the second values of the second line buffer in the destination buffer comprises to:
use a hardware instruction to interleave the first values and the second values in the destination buffer.
3. The apparatus of claim 2 , wherein the hardware instruction is one of unpackhi or unparcklo.
4. The apparatus of claim 2 , wherein the processor is further configured to:
calculate a respective cumulative sum for at least some locations of the destination buffer.
5. The apparatus of claim 4 , wherein to calculate the respective cumulative sum for the at least some locations of the destination buffer comprises to:
add a destination buffer value at a location of the at least some locations of the destination buffer to a cumulative sum corresponding to an immediately preceding location of the destination buffer.
6. The apparatus of claim 5 , wherein to determine, using the destination buffer, the probability distribution for coding the current value of the current scan-order anti-diagonal line comprises to:
calculate a difference between a first cumulative sum corresponding to a first location of the destination buffer and a second cumulative sum corresponding to a second location of the destination buffer, wherein the first cumulative sum and the second cumulative sum are selected based on a position of the current value; and
determine the probability distribution using the difference.
7. The apparatus of claim 6 , wherein a distance between the first location of the destination buffer and the second location of the destination buffer is related to a number of scan positions of already coded values of a template, wherein the template indicates, for a to-be-coded value, scan positions arranged, in the template, in at least two anti-diagonal lines.
8. An apparatus for decoding a transform block, the apparatus comprising:
a processor configured to:
store, in a first line buffer, first values of a first scan-order diagonal line scanned immediately before a current scan-order diagonal line of the transform block;
store, in a second line buffer, second values of a second scan-order diagonal line scanned immediately before the first scan-order diagonal line;
interleave the first values of the first line buffer and the second values of the second line buffer in a destination buffer;
select, using the destination buffer, a probability distribution for coding a current value of the current scan-order diagonal line;
entropy decode, from a compressed bitstream, the current value using the probability distribution; and
replace, for coding values of an immediately subsequent scan-order diagonal line to the current scan-order diagonal line, one of the second line buffer or the first line buffer with current values of the current scan-order diagonal line.
9. The apparatus of claim 8 , wherein to interleave the first values of the first line buffer and the second values of the second line buffer in the destination buffer comprises to:
use a hardware instruction to interleave the first values and the second values in the destination buffer.
10. The apparatus of claim 9 , wherein the hardware instruction is one of unpackhi or unparcklo.
11. The apparatus of claim 9 , wherein the processor is further configured to:
calculate a respective cumulative sum for at least some locations of the destination buffer.
12. The apparatus of claim 11 , wherein to calculate the respective cumulative sum for the at least some locations of the destination buffer comprises to:
add a destination buffer value at a location of the at least some locations of the destination buffer to a cumulative sum corresponding to an immediately preceding location of the destination buffer.
13. The apparatus of claim 12 , wherein to determine, using the destination buffer, the probability distribution for coding the current value of the current scan-order diagonal line comprises to:
calculate a difference between a first cumulative sum corresponding to a first location of the destination buffer and a second cumulative sum corresponding to a second location of the destination buffer, wherein the first cumulative sum and the second cumulative sum are selected based on a position of the current value; and
determine the probability distribution using the difference.
14. The apparatus of claim 13 , wherein a distance between the first location of the destination buffer and the second location of the destination buffer is related to a number of scan positions of already coded values of a template, wherein the template indicates, for a to-be-coded value, scan positions arranged, in the template, in at least two anti-diagonal lines.
15. A method for decoding a transform block, comprising:
storing, in a first line buffer, first values of a first scan-order diagonal line scanned immediately before a current scan-order diagonal line of the transform block;
storing, in a second line buffer, second values of a second scan-order diagonal line scanned immediately before the first scan-order diagonal line;
interleaving the first values of the first line buffer and the second values of the second line buffer in a destination buffer;
selecting, using the destination buffer, a probability distribution for coding a current value of the current scan-order diagonal line;
entropy decoding, from a compressed bitstream, the current value using the probability distribution; and
replacing, for coding values of an immediately subsequent scan-order diagonal line to the current scan-order diagonal line, one of the second line buffer or the first line buffer with current values of the current scan-order diagonal line.
16. The method of claim 15 , wherein interleaving the first values of the first line buffer and the second values of the second line buffer in the destination buffer comprises:
using a hardware instruction to interleave the first values and the second values in the destination buffer.
17. The method of claim 16 , wherein the hardware instruction is one of unpackhi or unparcklo.
18. The method of claim 16 , further comprising:
calculating a respective cumulative sum for at least some locations of the destination buffer.
19. The method of claim 18 , wherein calculating the respective cumulative sum for the at least some locations of the destination buffer comprises:
adding a destination buffer value at a location of the at least some locations of the destination buffer to a cumulative sum corresponding to an immediately preceding location of the destination buffer.
20. The method of claim 19 , wherein determining, using the destination buffer, the probability distribution for coding the current value of the current scan-order diagonal line comprises:
calculating a difference between a first cumulative sum corresponding to a first location of the destination buffer and a second cumulative sum corresponding to a second location of the destination buffer, wherein the first cumulative sum and the second cumulative sum are selected based on a position of the current value; and
determining the probability distribution using the difference.