Ascertaining audience reactions for a media item
AssigneeIMDb.com, Inc.
InventorJamie Russell Hunter Crawley
Life of this patent
Estimating the term…
Abstract
Disclosed are various embodiments for identifying and classifying audience reactions during a playback of a media item. In one embodiment, among others, a computing device is used to identify a reaction event from sensor data of a participant consuming a playback of a media item. The reaction event is correlated with a point in time of the media item. A reaction classification for the reaction event is determined based on the sensor data, and the reaction classification is associated with the point in time of the media item.
Claims20
A non-transitory computer-readable medium embodying a program executable in at least one computing device, wherein the program, when executed, causes the at least one computing device to at least:
- identify a playback of a media item in a media environment for a plurality of participants;
- identify in a video signal a plurality of motion events individually associated with the plurality of participants viewing the media item, the video signal being captured by an image capture device associated with the media environment, the plurality of motion events being associated with a point in time of the media item, the plurality of motion events occurring within a time period from the point in time;
- determine that a quantity of the plurality of participants causing a subset of the plurality of motion events meets a threshold number of participants in the media environment, the subset of the plurality of motion events having a respective time stamp captured within a time threshold from the point in time of the playback of the media item;
- determine a reaction classification based at least in part on a comparison between the plurality of motion events and a sequence of transitions for a plurality of reference points for a model expression, wherein the plurality of reference points comprises at least one of a facial reference point or a body reference point;
- and
- associate the reaction classification with the point in time of the media item.
A system, comprising:
- at least one computing device;
- and
- an application executable in the at least one computing device, wherein the application, when executed, causes the at least one computing device to at least:
- identify a plurality of reaction events from sensor data generated by at least one participant consuming a playback of a media item;
- correlate the plurality of reaction events with a point in time of the media item;
- determine a reaction classification for the plurality of reaction events based at last in part on a comparison between the plurality of reaction events and a sequence of transitions for a plurality of reference points for a model expression;
- and
- associate the reaction classification with the point in time of the media item.
A method, comprising:
- identifying, via at least one computing device, in a video signal from a plurality of reaction events individually associated with a plurality of participants viewing a playback of a media item, the video signal being captured by an image capture device;
- correlating, via the at least one computing device, the plurality of reaction events with a point in time of the media item based at least in part on an audio signal of the media item;
- determining, via the at least one computing device, a reaction classification for the plurality of reaction events based at least in part on a comparison between a video segment of the plurality of reaction events and a sequence of transitions for a plurality of reference points for a model expression, wherein the plurality of reference points comprises at least one of a facial reference point or a body reference point;
- and
- associating, via the at least one computing device, the reaction classification with the point in time of the media item.
The method of claim of 12, wherein the video signal is captured by at least one of:
- a thermal image camera, a point-cloud camera, or a three-dimensional scanner, and wherein identifying in the video signal the plurality of reaction events is based at least in part on a limb movement of at least one of the plurality of participants in the video signal.
Dependencies are inferred from the claim text.