Coefficient context modeling in video coding
In some embodiments, a method determines a plurality of classes of bins that are used to determine a context model for entropy coding of a current block in a video. The method calculates a first value for a first class of bins in the plurality of classes of bins and calculates a second value for a second class of bins in the plurality of classes of bins. The first value for the first class of bins is weighted by a first weight to generate a weighted first value and the second value for the second class of bins is weighted by a second weight to generate a weighted second value. The method then selects a context model based on the first weighted value and the second weighted value.
1. A method comprising:
determining, by a computing device, a plurality of classes of bins that are used to determine a context model for entropy coding of a current block in a video;
calculating, by the computing device, a first value for a first class of bins in the plurality of classes of bins for a first bin plane;
calculating, by the computing device, a second value for a second class of bins in the plurality of classes of bins for the first bin plane;
weighting, by the computing device, the first value for the first class of bins by a first weight to generate a weighted first value and weighting the second value for the second class of bins by a second weight to generate a weighted second value;
selecting, by the computing device, a third value for a second bin plane; and
selecting, by the computing device, a context model based on the first weighted value, the second weighted value, and the third value.
2. The method of claim 1 , wherein a first center point in a first bin in the first class of bins is closer in distance to the current block compared to a second center point in a second bin in the second class of bins.
3. The method of claim 1 , wherein center points in bins in the first class of bins are closer in distance to the current block compared to center points in bins in the second class of bins.
4. The method of claim 1 , wherein bins in the first class of bins are positioned in a first line and bins in the second class of bins are positioned in a second line.
5. The method of claim 1 , wherein the first class of bins and the second class of bins are in a local template for the current block.
6. The method of claim 1 , wherein the first weight weights the first class of bins higher than the second weight for the second class of bins.
7. The method of claim 1 , wherein selecting the context model comprises:
selecting a content index based on the first weighted value, the second weighted value, and the third value, wherein the content index points to the context model.
8. The method of claim 1 , wherein:
calculating the first value for a first class of bins in the plurality of classes of bins comprises calculating a first sum based on coefficient values for bins in the first class of bins; and
calculating the second value for the second class of bins in the plurality of classes of bins comprises calculating a second sum based on coefficient values for bins in the second class of bins.
9. The method of claim 8 , wherein:
the first sum is a first number of bins with non-zero coefficient values in the first class of bins, and
the second sum is a second number of bins with non-zero coefficient values in the second class of bins.
10. The method of claim 8 , wherein:
weighting the first value for the first class of bins comprises weighting the first sum by the first weight to generate a first weighted sum, and
weighting the second value for the second class of bins comprises weighting the second sum by the second weigh to generate a second weighted sum.
11. The method of claim 10 , wherein selecting the context model comprises using the first weighted sum, the second weighted sum, and the third value to determine the context model.
12. The method of claim 1 , wherein determining the plurality of classes of bins comprises adaptively determining a number of the plurality of classes based on characteristics of the video.
13. The method of claim 12 , wherein the characteristics comprise one or more of a resolution of the video, a quantization parameter, and a number of non-zero coefficients in a template of bins for the current block.
14. The method of claim 1 , wherein determining the plurality of classes of bins comprises adaptively determining a number of the context models based on characteristics of the video.
15. A non-transitory computer-readable storage medium containing instructions, that when executed, control a computer system to be operable for:
determining a plurality of classes of bins that are used to determine a context model for entropy coding of a current block in a video;
calculating a first value for a first class of bins in the plurality of classes of bins for a first bin plane;
calculating a second value for a second class of bins in the plurality of classes of bins for the first bin plane;
weighting the first value for the first class of bins by a first weight to generate a weighted first value and weighting the second value for the second class of bins by a second weight to generate a weighted second value;
selecting a third value for a second bin plane; and
selecting a context model based on the first weighted value, the second weighted value, and the third value.
16. The non-transitory computer-readable storage medium of claim 15 , wherein a first center point in a first bin in the first class of bins is closer in distance to the current block compared to a second center point in a second bin in the second class of bins.
17. The non-transitory computer-readable storage medium of claim 15 , wherein center points in bins in the first class of bins are closer in distance to the current block compared to center points in bins in the second class of bins.
18. The non-transitory computer-readable storage medium of claim 15 , wherein bins in the first class of bins are positioned in a first line and bins in the second class of bins are positioned in a second line.
19. The non-transitory computer-readable storage medium of claim 15 , wherein:
calculating the first value for a first class of bins in the plurality of classes of bins comprises calculating a first sum based on coefficient values for bins in the first class of bins; and
calculating the second value for the second class of bins in the plurality of classes of bins comprises calculating a second sum based on coefficient values for bins in the second class of bins.
20. An apparatus comprising:
one or more computer processors; and
a non-transitory computer-readable storage medium comprising instructions, that when executed, control the one or more computer processors to be operable for:
determining a plurality of classes of bins that are used to determine a context model for entropy coding of a current block in a video;
calculating a first value for a first class of bins in the plurality of classes of bins for a first bin plane;
calculating a second value for a second class of bins in the plurality of classes of bins for the first bin plane;
weighting the first value for the first class of bins by a first weight to generate a weighted first value and weighting the second value for the second class of bins by a second weight to generate a weighted second value;
selecting a third value for a second bin plane; and
selecting a context model based on the first weighted value, the second weighted value, and the third value.