Higher order manifest data compression

Patent metadata

Field	Value
Publication number	US-11063997-B1
Application number	US-201916367651-A
Country	US
Kind code	B1
Filing date	Mar 28, 2019
Priority date	Mar 28, 2019
Publication date	Jul 13, 2021
Grant date	Jul 13, 2021

Abstract

Methods and apparatus are described for efficiently representing manifests and playlists for media presentations. Some media streaming protocols include repetition syntax by which a sequence of media segments having the same duration can be represented with a single entry. A pattern syntax is introduced by which repeating patterns of segment durations in a cycle may be represented in manifests and playlists.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving a media presentation, the media presentation including a sequence of video segments and a sequence of audio segments, the video segments all being of a same duration, the sequence of audio segments including first audio segments of a first duration, and second audio segments of a second duration different from the first duration, the sequence of audio segments including a repeating pattern of the first and second audio segments;
generating video timeline data according to the dynamic adaptive streaming over HTTP (DASH) protocol, the video timeline data representing each of the video segments relative to a media timeline of the media presentation, the video timeline data employing an @r syntax of the DASH protocol to represent consecutive video segments having the same duration;
generating audio timeline data according to the DASH protocol, the audio timeline data representing each of the audio segments relative to the media timeline, the audio timeline data employing the @r syntax to represent consecutive audio segments having the first duration, the audio timeline data also including a pattern syntax that represents instances of the repeating pattern of the first and second audio segments; and
publishing the video timeline data and the audio timeline data as part of a DASH manifest configured to enable a client device to download or stream the media presentation.

2. The method of claim 1 , further comprising detecting the repeating pattern during generation of the audio timeline data by inspecting the sequence of audio segments.

3. The method of claim 1 , further comprising receiving the repeating pattern prior to generation of the audio timeline data.

4. The method of claim 1 , wherein the repeating pattern includes a plurality of the first audio segments followed by one or more of the second audio segments.

5. A computer-implemented method, comprising:
receiving a media presentation, the media presentation including a sequence of media segments, the sequence of media segments having a plurality of different durations, the sequence of media segments including a repeating pattern of the different durations;
generating media timeline data, the media timeline data representing each of the media segments relative to a media timeline of the media presentation, the media timeline data employing a pattern syntax to represent instances of the repeating pattern; and
publishing the media timeline data as part of a manifest configured to enable a client device to download or stream the media presentation.

6. The method of claim 5 , wherein the manifest is conformant with one of the dynamic adaptive streaming over HTTP (DASH protocol), the HTTP Live Streaming (HLS) protocol, the HTTP Dynamic Streaming protocol, or the Smooth Streaming protocol.

7. The method of claim 5 , wherein the pattern syntax in the manifest includes a reference to a pattern definition stored in metadata associated with the media presentation.

8. The method of claim 5 , further comprising detecting the repeating pattern during generation of the media timeline data by inspecting the sequence of media segments.

9. The method of claim 5 , further comprising receiving the repeating pattern prior to generation of the media timeline data.

10. The method of claim 5 , further comprising defining the repeating pattern during a fragmentation stage or a content encoding stage of an encoding and packaging pipeline.

11. The method of claim 5 , wherein the pattern syntax employs a pattern attribute, a plurality of duration attributes, and one or more repetition attributes, and wherein a segment element specifies values for each of the pattern attribute, the duration attributes, and the one or more repetition attributes to represent the instances of the repeating pattern.

12. The method of claim 5 , wherein the pattern syntax employs a pattern element, and wherein at least one of a plurality of segment elements includes a repetition attribute, each segment element including a duration attribute, each segment element also including a group attribute that represents a corresponding group to which the segment element belongs, a first group of the segment elements corresponding to one or more instances of the repeating pattern, the pattern element including the group attribute representing the first group, the pattern element also including the repetition attribute representing a number of the instances of the repeating pattern.

13. The method of claim 5 , wherein the pattern syntax employs a pattern element, and wherein at least one of a plurality of segment elements includes a repetition attribute, each segment element including a duration attribute, a subset of the segment elements being child elements of the pattern element, the subset of the segment elements corresponding to one or more instances of the repeating pattern, the pattern element including the repetition attribute representing a number of the instances of the repeating pattern.

14. A system, comprising one or more computing devices configured to:
receive a media presentation, the media presentation including a sequence of media segments, the sequence of media segments having a plurality of different durations, the sequence of media segments including a repeating pattern of the different durations;
generate media timeline data, the media timeline data representing each of the media segments relative to a media timeline of the media presentation, the media timeline data employing a pattern syntax to represent instances of the repeating pattern; and
publish the media timeline data as part of a manifest configured to enable a client device to download or stream the media presentation.

15. The system of claim 14 , wherein the manifest is conformant with one of the dynamic adaptive streaming over HTTP (DASH protocol), the HTTP Live Streaming (HLS) protocol, the HTTP Dynamic Streaming protocol, or the Smooth Streaming protocol.

16. The system of claim 14 , wherein the pattern syntax in the manifest includes a reference to a pattern definition stored in metadata associated with the media presentation.

17. The system of claim 14 , wherein the one or more computing devices are further configured to detect the repeating pattern during generation of the media timeline data by inspecting the sequence of media segments.

18. The system of claim 14 , wherein the one or more computing devices are further configured to receive the repeating pattern prior to generation of the media timeline data.

19. The system of claim 14 , wherein the one or more computing devices are further configured to define the repeating pattern during a fragmentation stage or a content encoding stage of an encoding and packaging pipeline.

20. The system of claim 14 , wherein the pattern syntax employs a pattern attribute, a plurality of duration attributes, and one or more repetition attributes, and wherein a segment element specifies values for each of the pattern attribute, the duration attributes, and the one or more repetition attributes to represent the instances of the repeating pattern.

21. The system of claim 14 , wherein the pattern syntax employs a pattern element, and wherein at least one of a plurality of segment elements includes a repetition attribute, each segment element including a duration attribute, each segment element also including a group attribute that represents a corresponding group to which the segment element belongs, a first group o

Assignees

• Amazon Tech Inc

Inventors

• Pedrini Mairo
• Koceski Kyle Bradley
• Lamb Joshua
• Wu Yongjun
• Parminder Singh
• Coleman Michael
• Weil Nicolas

Classifications

• H04L65/612

  for unicast · CPC title

• H04L65/75

  Media network packet handling · CPC title

• H04N21/8456 Primary

  by decomposing the content in the time domain, e.g. in time segments · CPC title

• H04L65/70

  Media network packetisation · CPC title

• H04L65/65

  Network streaming protocols, e.g. real-time transport protocol [RTP] or real-time control protocol [RTCP] · CPC title