System and method for automatic generation of sports media highlights

• US 10,417,500 B2
• Filed: 12/28/2017
• Issued: 09/17/2019
• Est. Priority Date: 12/28/2017
• Status: Active Grant

• Current Assignee: Disney Enterprises Incorporated (The Walt Disney Company)
• Original Assignee: Disney Enterprises Incorporated (The Walt Disney Company)
• Inventors: Ray, Abhik, Branch, Joel
• Primary Examiner(s): Chen, Cai Y
• Application Number:
  US15/857,142
• Publication Number:
  US 20190205652A1
• Time in Patent Office: 628 Days
• Field of Search: None
• CPC Class Codes: G06V 20/42 of sport video content
• CPC Class Codes: G06V 20/44 Event detection
• CPC Class Codes: G06V 20/47 Detecting features for summarising video content

Abstract

Methods and systems for automatic generation of sports highlights are described, where AV media data of a sporting event (game) are analyzed to identify sports highlights using a data-graph-based technique for “in-game” highlights, and “out-of-game” AV media data (e.g. post-game commentary/analysis/news) are analyzed for out-of-game content (OGC) related to each of the selected in-game highlights to produce a collective AV narrative around each of the in-game highlights selected. The format and appearance of the final highlights AV media content or video are also adjustable by the user or the highlights producer to optimize the highlight viewing experience.

Claims

1. A method for automatically generating a narrative based highlights video of a sporting event, comprising:

receiving AV media data indicative of the sporting event;

identifying at least one in-game highlight within the AV media data using a data-graph technique as selected in-game highlight, the data-graph technique being performed by a computer other than a user device;

wherein the data-graph technique uses locations and actions of the players and possession during the sporting event, and wherein the locations and actions and possession are represented by data indicative of nodes and edges of a data graph;

identifying out-of-game content related to the selected in-game highlight as selected out-of-game content; and

combining the selected in-game highlight with the selected out-of-game content to create the narrative-based sports highlight video.

2. The method of claim 1, wherein the graph-based technique comprises generating dynamic data graphs for plays in the sporting event.

3. The method of claim 1, wherein the graph-based technique comprises identifying frequent sub-graphs for plays in the sporting event.

4. The method of claim 1, wherein the graph-based technique comprises identifying dynamic normative patterns for the sporting event.

5. The method of claim 3, wherein the graph-based technique further comprises identifying dynamic normative patterns for the sporting event from the frequent sub-graphs.

6. The method of claim 4, wherein the graph-based technique further comprises identifying dynamic normative patterns for the current sporting event closest to dynamic normative patterns of other sporting events with the same outcome.

7. The method of claim 1, wherein the identifying out-of-game content comprises creating out-of-game content AV clips based on topic.

8. The method of claim 7, wherein the creating the clips comprises segmenting AV content of out-of-game sports shows into segments of a predetermined length, analyzing the topic of each segment sequentially, and identifying the segment when a topic change occurs.

9. The method of claim 1, wherein the identifying further comprises creating an in-game highlight video having an AV clip of at least one highlight selected using the graph-based technique and an AV clip that is a traditional highlight.

10. The method of claim 1 wherein at least one play in the selected in-game highlight is weighted by at least one of pre-game commentary and historical data regarding the teams in the sporting event.

11. The method of claim 1, wherein the identifying at least one in-game highlight, the identifying out-of-game content, and the combining are adjusted based on at least one of feedback data and user preferences.

12. The method of claim 1, wherein the combining is based on at least one of rules, weightings and preferences.

13. The method of claim 1, wherein identifying at least one in-game highlight, the identifying out-of-game content, and the combining are based on weighting factors.

14. The method of claim 13, wherein the weighting factors comprises at least one of play weighting factors and out-of-game content (OGC) weighting factors.

15. The method of claim 14, wherein a value of the weighting factor is based on events that happen after a “

point of no return”

in the sporting event.

25. The method of claim 1, wherein the nodes are indicative of player locations and possession and the edges are indicative of player actions.

27. The method of claim 1, further comprising receiving out-of-game AV media data and the identifying comprises identifying, from the out-of-game AV media data, out-of-game content related to the selected in-game highlight as the selected out-of-game content.

16. A method for automatically generating a highlights video of a sporting event, comprising:

receiving AV media data indicative of the sporting event;

identifying at least one in-game highlight within the AV media data using a data-graph technique as selected in-game highlight, the data-graph technique being performed by a computer other than a user device;

wherein the data-graph technique uses locations and actions of the players and possession during the sporting event, and wherein the locations and actions and possession are represented by data indicative of nodes and edges of a data graph;

wherein the graph-based technique comprises identifying at least one dynamic normative pattern for the sporting event; and

providing the sports highlight video using the selected in-game highlight.

17. The method of claim 16, wherein the graph-based technique further comprises identifying frequent sub-graphs for plays in the sporting event, and wherein the dynamic normative pattern is identified from the frequent sub-graphs.

18. The method of claim 16, wherein the at least one dynamic normative pattern is related to the outcome of the sports event.

19. The method of claim 16, further comprising identifying out-of-game content related to the selected in-game highlight as selected out-of-game content;

and combining the selected in-game highlight with the selected out-of-game content to create the sports highlight video.

26. The method of claim 16, wherein the nodes are indicative of player locations and possession and the edges are indicative of player actions.

20. A method for automatically generating a narrative based highlights video of a sports event, comprising:

receiving AV media data indicative of the sporting event;

identifying at least two in-game highlights within the AV media data, a first in-game highlight using a traditional in-game highlight, and a second in-game highlight using a data graph technique, as selected in-game highlights, the data-graph technique being performed by a computer other than a user device;

wherein the data-graph technique uses locations and actions of the players and possession during the sporting event, and wherein the locations and actions and possession are represented by data indicative of nodes and edges of a data graph;

identifying out-of-game content related to the selected in-game highlight as selected out-of-game content; and

combining the selected in-game highlights with the out-of-game content to create the narrative-based sports highlight video.

21. The method of claim 20, wherein the identifying out-of-game content comprises creating out-of-game content AV clips based on topic.

22. The method of claim 20, wherein the creating the clips, comprises segmenting AV content of out-of-game sports shows into segments of a predetermined length, analyzing each segment sequentially, and identifying when a topic change occurs by marking the segment.

23. The method of claim 20, wherein the combining is based on at least one of rules, weightings and preferences.

24. The method of claim 20, wherein at least one of the out-of-game content shows, commentators, number of clips, and types, are selectable to be part of the highlights video.

28. The method of claim 20, further comprising receiving out-of-game AV media data and the identifying comprises identifying, from the out-of-game AV media data, out-of-game content related to the selected in-game highlight as the selected out-of-game content.