Techniques for efficiently performing subsequence-based encoding for a media title

Patent metadata

Field	Value
Publication number	US-11425432-B2
Application number	US-202117328358-A
Country	US
Kind code	B2
Filing date	May 24, 2021
Priority date	Feb 4, 2019
Publication date	Aug 23, 2022
Grant date	Aug 23, 2022

Abstract

In various embodiments, an interpolation-based encoding application encodes a first subsequence included in a media title at each encoding point included in a first set of encoding points to generate encoded subsequences. Subsequently, the interpolation-based encoding application performs interpolation operation(s) based on the encoded subsequences to estimate a first media metric value associated with a first encoding point that is not included in the first set of encoding points. The interpolation-based encoding application then generates an encoding recipe based on the encoded subsequences and the first media metric value. The encoding recipe specifies a different encoding point for each subsequence included in the media title. After determining that the encoding recipe specifies the first encoding point for the first subsequence, the interpolation-based encoding application encodes the first subsequence at the first encoding point to generate at least a portion of an encoded version of the media title.

First claim

What is claimed is:

1. A computer-implemented method for encoding a media title, the method comprising:
performing one or more interpolation operations based on a first plurality of encoded subsequences corresponding to a first subsequence included in the media title to estimate a first plurality of media metric values;
generating a first encoding recipe that specifies a first encoding point for the first subsequence based on the first plurality of media metric values; and
encoding the first subsequence based on the first encoding recipe.

2. The method of claim 1 , further comprising encoding the first subsequence across a first plurality of encoding points used to generate the first plurality of encoded subsequences.

3. The method of claim 1 , wherein the first encoding point is not included in a first plurality of encoding points used to generate the first plurality of encoded subsequences.

4. The method of claim 1 , wherein the first plurality of media metric values comprise a value for a bitrate, a size, a peak signal-to-noise-ratio, a linear video multimethod assessment fusion metric, or a harmonic video multimethod assessment fusion.

5. The method of claim 1 , wherein each encoding point included in a first plurality of encoding points used to generate the first plurality of encoded subsequences is associated with a different value of an encoding parameter.

6. The method of claim 1 , wherein the first plurality of media metric values comprise a first value of a visual quality metric, and further comprising:
decoding the first plurality of encoded subsequences to generate a plurality of reconstructed subsequences;
computing a plurality of values for the visual quality metric based on the plurality of reconstructed subsequences; and
interpolating between two values included in the plurality of values for the visual quality metric to estimate the first value of the visual quality metric.

7. The method of claim 1 , wherein the first plurality of media metric values comprises a first bitrate, and further comprising:
determining a plurality of bitrates based of the first plurality of encoded subsequences; and
interpolating between two bitrates included in the plurality of bitrates to estimate the first bitrate.

8. The method of claim 1 , wherein generating the first encoding recipe comprises:
performing the one or more interpolation operations based on the first plurality of encoded subsequences to estimate a first bitrate associated with the first encoding point;
performing one or more optimization operations based on the first plurality of encoded subsequences, the first plurality of media metric values, and the first bitrate to generate a plurality of encoding recipes; and
determining the first encoding recipe from the plurality of encoding recipes based on a first target bitrate or a first target value for a visual quality metric.

9. The method of claim 1 , wherein generating the first encoding recipe comprises:
generating a first convex hull of rate-quality points based on the first plurality of encoded subsequences and the first plurality of media metric values;
generating a global convex hull of encoding recipes based on the first convex hull and a second convex hull that is associated with a second subsequence included in the media title; and
determining the first encoding recipe from the global convex hull based on a target bitrate or a target level of visual quality.

10. The method of claim 1 , wherein the first encoding recipe specifies a different encoding point for each subsequence included in the media title.

11. One or more non-transitory computer-readable media including instructions that, when executed by one or more processors, cause the one or more processors to encode a media title by performing the steps of:
performing one or more interpolation operations based on a first plurality of encoded subsequences corresponding to a first subsequence included in the media title to estimate a first plurality of media metric values;
selecting a first encoding point for the first subsequence based on the first plurality of media metric values;
and
encoding the first subsequence at the first encoding point.

12. The one or more non-transitory computer-readable media of claim 11 , further comprising encoding the first subsequence across a first plurality of encoding points to generate the first plurality of encoded subsequences.

13. The one or more non-transitory computer-readable media of claim 11 , wherein the first encoding point is not included in a first plurality of encoding points used to generate the first plurality of encoded subsequences.

14. The one or more non-transitory computer-readable media of claim 11 wherein the first plurality of media metric values comprise a value for a bitrate, a size, a peak signal-to-noise-ratio, a linear video multimethod assessment fusion metric, or a harmonic video multimethod assessment fusion.

15. The one or more non-transitory computer-readable media of claim 11 , wherein each encoding point included in a first plurality of encoding points used to generate the first plurality of encoded subsequences is associated with a different value of an encoding parameter.

16. The one or more non-transitory computer-readable media of claim 11 , wherein the first plurality of media metric values comprise a first value of a visual quality metric, and further comprising:
decoding the first plurality of encoded subsequences to generate a plurality of reconstructed subsequences;
computing a plurality of values for the visual quality metric based on the plurality of reconstructed subsequences; and
interpolating between two values included in the plurality of values for the visual quality metric to estimate the first value of the visual quality metric.

17. The one or more non-transitory computer-readable media of claim 11 , wherein the first plurality of media metric values comprise a first bitrate, and further comprising:
determining a plurality of bitrates based of the first plurality of encoded subsequences; and
interpolating between two bitrates included in the plurality of bitrates to estimate the first bitrate.

18. The one or more non-transitory computer-readable media of claim 11 , wherein selecting the first encoding point for the first subsequence comprises:
performing the one or more interpolation operations based on the first plurality of encoded subsequences to estimate a first bitrate associated with the first encoding point;
performing one or more optimization operations based on the first plurality of encoded subsequences, the first plurality of media metric values, and the first bitrate to generate a plurality of encoding recipes; and
determining a first encoding recipe from the plurality of encoding recipes based on a first target bitrate or a first target value for a visual quality metric, wherein the first encoding recipe specifies the first encoding point for the first subsequence.

19. The one or more non-transitory computer-readable media of claim 11 , wherein selecting the first encoding point for the first subsequence comprises:
generating a first convex hull of rate-quality points based on the first plurality of encoded subsequences and the first plurality of media metric values;
generating a global convex hull of encoding recipes based on the first convex hull and a second convex hull that is associated with a second subsequence included in the media title; and
determining a first encoding recipe from the global convex hull based on a target

Assignees

• Netflix Inc

Inventors

• Wallendael Glenn Van
• Aaron Anne
• Swanson Kyle
• De Cock Jan
• Guo Liwei
• Bhaskar Sonia

Classifications

• H04N21/23439 Primary

  for generating different versions · CPC title

• H04N21/234354

  by altering signal-to-noise ratio parameters, e.g. requantization · CPC title

• H04L65/70

  Media network packetisation · CPC title

• H04N19/147 Primary

  according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567) · CPC title

• H04L65/765

  intermediate · CPC title