Scene and activity identification in video summary generation
First Claim
1. A method of generating a video summary of a video, the method comprising:
- accessing metadata associated with a video, the accessed metadata representative of one or more aspects of the capture of the video as a function of time during capture of the video;
identifying patterns in the metadata as a function of time that correspond to performance of one or more activities being performed by a subject of the video;
determining the one or more activities being performed by the subject of the video during specific portions of the video based on the identifications of the patterns in the metadata as a function of time, the one or more activities including a first activity performed by the subject of the video during a first portion of the video, the first activity being of a given type of activity;
identifying moments within the video at which events of interest are captured in the video based on the accessed metadata, the moments including a first moment during performance of the first activity by the subject of the video at which a first event of interest occurs, the first event being of a given type of event;
identifying individual highlight scenes in the video for the individual events of interest, wherein lengths of footage in the video included in the highlight scenes before and after the moments at which the events of interest occur are based on types of the activities being performed by the subject of the video at the moments in the video at which the events of interest are captured and types of the events such that a first scene in the video is identified for the first event of interest and a length of footage included in the first scene before and after the first moment is a first length based on the given type of activity being of a first type of activity and the given type of event being of a first type of event, a second length based on the given type of activity being of the first type of activity and the given type of event being of a second type of event, a third length based on the given type of activity being of a second type of activity and the given type of event being of the first type of event, and a fourth length based on the given type of activity being of the second type of activity and the given type of event being of the second type of event, the first length different from the second length and the third length, the third length different from the fourth length; and
generating a video summary of the video for playback, the video summary including at least one of the highlight scenes.
5 Assignments
0 Petitions
Accused Products
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.
262 Citations
18 Claims
-
1. A method of generating a video summary of a video, the method comprising:
-
accessing metadata associated with a video, the accessed metadata representative of one or more aspects of the capture of the video as a function of time during capture of the video; identifying patterns in the metadata as a function of time that correspond to performance of one or more activities being performed by a subject of the video; determining the one or more activities being performed by the subject of the video during specific portions of the video based on the identifications of the patterns in the metadata as a function of time, the one or more activities including a first activity performed by the subject of the video during a first portion of the video, the first activity being of a given type of activity; identifying moments within the video at which events of interest are captured in the video based on the accessed metadata, the moments including a first moment during performance of the first activity by the subject of the video at which a first event of interest occurs, the first event being of a given type of event; identifying individual highlight scenes in the video for the individual events of interest, wherein lengths of footage in the video included in the highlight scenes before and after the moments at which the events of interest occur are based on types of the activities being performed by the subject of the video at the moments in the video at which the events of interest are captured and types of the events such that a first scene in the video is identified for the first event of interest and a length of footage included in the first scene before and after the first moment is a first length based on the given type of activity being of a first type of activity and the given type of event being of a first type of event, a second length based on the given type of activity being of the first type of activity and the given type of event being of a second type of event, a third length based on the given type of activity being of a second type of activity and the given type of event being of the first type of event, and a fourth length based on the given type of activity being of the second type of activity and the given type of event being of the second type of event, the first length different from the second length and the third length, the third length different from the fourth length; and generating a video summary of the video for playback, the video summary including at least one of the highlight scenes. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A system that generates a video summary of a video, the system comprising:
-
a non-transitory computer-readable storage medium storing instructions configured to, when executed; access metadata associated with a video, the accessed metadata representative of one or more aspects of capture of the video as a function of time during capture of the video; identify patterns in the metadata as a function of time that correspond to performance of one or more activities being performed by a subject of the video; determine the one or more activities being performed by the subject of the video during specific portions of the video based on the identifications of the patterns in the metadata as a function of time, the one or more activities including a first activity performed by the subject of the video during a first portion of the video, the first activity being of a given type of activity; identify moments within the video at which events of interest are captured in within the video based on the accessed metadata, the moments including a first moment during performance of the first activity by the subject of the video at which a first event of interest occurs, the first event being of a given type of event; identify individual highlight scenes in the video for the individual events of interest, wherein lengths of footage in the video included in the highlight scenes before and after the moments at which the events of interest occur are based on types of the activities being performed by the subject of the video at the moments in the video at which the events of interest are captured and types of the events such that a first scene in the video is identified for the first event of interest and a length of footage included in the first scene before and after the first moment is a first length based on the given type of activity being of a first type of activity and the given type of event being of a first type of event, a second length based on the given type of activity being of the first type of activity and the given type of event being of a second type of event, a third length based on the given type of activity being of a second type of activity and the given type of event being of the first type of event, and a fourth length based on the given type of activity being of the second type of activity and the given type of event being of the second type of event, the first length different from the second length and the third length, the third length different from the fourth length; and generate a video summary of the video for playback, the video summary including at least one of the highlight scenes; and a processor configured to execute the instructions. - View Dependent Claims (12, 13, 14)
-
-
15. A non-transitory computer-readable storage medium storing instructions for identifying scenes in captured video for inclusion in a video summary, the instructions configured to, when executed:
-
access metadata associated with a video, the accessed metadata representative of one or more aspects of capture of the video as a function of time during capture of the video; identify patterns in the metadata as a function of time that correspond to performance of one or more activities being performed by a subject of the video; determine the one or more activities being performed by the subject of the video during specific portions of the video based on the identifications of the patterns in the metadata as a function of time, the one or more activities including a first activity performed by the subject of the video during a first portion of the video, the first activity being of a given type of activity; identify moments within the video at which events of interest are captured in within the video based on the accessed metadata, the moments including a first moment during performance of the first activity by the subject of the video at which a first event of interest occurs, the first event being of a given type of event; identify individual highlight scenes in the video for the individual events of interest, wherein lengths of footage in the video included in the highlight scenes before and after the moments at which the events of interest occur are based on types of the activities being performed by the subject of the video at the moments in the video at which the events of interest are captured and types of the events such that a first scene in the video is identified for the first event of interest and a length of footage included in the first scene before and after the first moment is a first length based on the given type of activity being of a first type of activity and the given type of event being of a first type of event, a second length based on the given type of activity being of the first of activity and the given type of event being of a second type of event, a third length based on the given type of activity being of a second type of activity and the given type of event being of the first type of event, and a fourth length based on the given type of activity being of the second type of activity and the given type of event being of the second type of event, the first length different from the second length and the third length, the third length different from the fourth length; and generate a video summary of the video for playback, the video summary including at least one of the highlight scenes. - View Dependent Claims (16, 17, 18)
-
Specification