Scene and activity identification in video summary generation

Patent metadata

Field	Value
Publication number	US-11776579-B2
Application number	US-202117378324-A
Country	US
Kind code	B2
Filing date	Jul 16, 2021
Priority date	Jul 23, 2014
Publication date	Oct 3, 2023
Grant date	Oct 3, 2023

Abstract

Video and corresponding metadata is accessed. Events of interest within the video are identified based on the corresponding metadata, and best scenes are identified based on the identified events of interest. A video summary can be generated including one or more of the identified best scenes. The video summary can be generated using a video summary template with slots corresponding to video clips selected from among sets of candidate video clips. Best scenes can also be identified by receiving an indication of an event of interest within video from a user during the capture of the video. Metadata patterns representing activities identified within video clips can be identified within other videos, which can subsequently be associated with the identified activities.

First claim

What is claimed is:

1. A system configured to generate a video summary, the system comprising:
one or more physical processors configured by machine readable instructions to:
obtain metadata for one or more videos;
analyze the metadata for the one or more videos to identify events of interest in the one or more videos;
identify video portions of the one or more videos as candidates for inclusion in the video summary based on the events of interest in the one or more videos, the video summary being characterized b a given length, the identified video portions including a first video portion and a second video portion;
rank the identified video portions, wherein the first video portion is ranked higher than the second video portion; and
select among the identified video portions for inclusion in the video summary based on rankings of the identified video portions, wherein the first video portion is selected over the second video portion for inclusion in the video summary based on the first video portion being ranked higher than the second video portion; and
generate the video summary from the selected video portions.

2. The system of claim 1 , wherein:
the metadata for the one or more videos includes location information identifying capture locations for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture location for the one or more videos being within a threshold distance of a pre-determined location; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold distance of the pre-determined location.

3. The system of claim 1 , wherein:
the metadata for the one or more videos includes time stamps representing capture times for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture time of the one or more videos being within a threshold time of a pre-determined time; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold time of the pre-determined time.

4. The system of claim 1 , wherein:
the metadata for the one or more videos includes location information identifying capture locations for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture location for the one or more videos being within a threshold distance of a user; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold distance of the user.

5. The system of claim 1 , wherein the identified video portions are ranked based on lengths of the identified video portions, with longer video portions being ranked higher than shorter video portions.

6. The system of claim 1 , wherein the identified video portions are ranked based on types of the metadata used to identify the events of interest.

7. The system of claim 6 , wherein the first video portion is ranked higher than the second video portion based on the first video portion being associated with a first event of interest identified based on image capture device velocity and the second video portion being associated with a second event of interest identified based on user heart rate.

8. The system of claim 1 , wherein the identified video portions are ranked based on activities associated with the identified video portions.

9. The system of claim 8 , wherein the first video portion is ranked higher than the second video portion based on the first video portion being associated with a jump or a crash and the second video portion being associated with sitting down or walking.

10. A method for generating a video summary, the method performed by a computing system including one or more processors, the method comprising:
obtaining, by the computing system, metadata for one or more videos;
analyzing, by the computing system, the metadata for the one or more videos to identify events of interest in the one or more videos;
identifying by the computing system, video portions of the one or more videos as candidates for inclusion in the video summary based on the events of interest in the one or more videos, the video summary being characterized by a given length, the identified video portions including a first video portion and a second video portion;
ranking, by the computing system, the identified video portions, wherein the first video portion is ranked higher than the second video portion; and
selecting, by the computing system, among the identified video portions for inclusion in the video summary based on rankings of the identified video portions, wherein the first video portion is selected over the second video portion for inclusion in the video summary based on the first video portion being ranked higher than the second video portion; and
generating, by the computing system, the video summary from the selected video portions.

11. The method of claim 10 , wherein:
the metadata for the one or more videos includes location information identifying capture locations for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture location for the one or more videos being within a threshold distance of a pre-determined location; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold distance of the pre-determined location.

12. The method of claim 10 , wherein:
the metadata for the one or more videos includes time stamps representing capture times for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture time of the one or more videos being within a threshold time of a pre-determined time; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold time of the pre-determined time.

13. The method of claim 10 , wherein:
the metadata for the one or more videos includes location information identifying capture locations for the one or more videos;
one or more of the events of interests are identified in the one or more videos based on corresponding capture location for the one or more videos being within a threshold distance of a user; and
one or more of the video portions are identified from the one or more videos as having been captured within the threshold distance of the user.

14. The method of claim 10 , wherein the identified video portions are ranked based on lengths of the identified video portions, with longer video portions being ranked higher than shorter video portions.

15. The method of claim 10 , wherein the identified video portions are ranked based on types of the metadata used to identify the events of interest.

16. The method of claim 15 , wherein the first video portion is ranked higher than the second video portion based on the first video portion being associated with a first event of interest identified based on image capture device velocity and the second video portion being associated with a second event of interest identified based on user heart rate.

17. The method of claim 10 , wherein the identified video portions are ranked based on activities associated with the identified video portions.

18. The method of claim 17 , wherein the first video portion is ranked higher th

Assignees

• Gopro Inc

Inventors

• Newman David A
• Silver Adam

Classifications

• G11B27/10 Primary

  Indexing; Addressing; Timing or synchronising; Measuring tape travel · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title

• G06V20/47 Primary

  Detecting features for summarising video content · CPC title

• G10L15/063

  Training · CPC title