Estimating engagement of consumers of presented content
AssigneeMicrosoft Corporation
InventorsJavier Hernandez Rivera, Zicheng Liu, Geoff Hulten, Michael Conrad, Kyle Krum, David DeBarr, Zhengyou Zhang
Life of this patent
Estimating the term…
Abstract
Technologies described herein relate to estimating engagement of a person with respect to content being presented to the person. A sensor outputs a stream of data relating to the person as the person is consuming the content. At least one feature is extracted from the stream of data, and a level of engagement of the person is estimated based at least in part upon the at least one feature. A computing function is performed based upon the estimated level of engagement of the person.
Claims20
A computer-executable method, comprising:
- receiving a stream of data output by a sensor, the stream of data relating to a person who is being presented with viewable content;
- extracting a feature of the person from the stream of data;
- computing a value that is indicative of engagement of the person with respect to the viewable content as the viewable content is being presented to the person, the computing of the value based at least in part upon the feature of the person extracted from the stream of data, wherein the computing of the value that is indicative of the engagement of the person comprises:
- analyzing the feature of the person over a predefined time window, wherein the feature is subject to alteration;
- and
- computing the value that is indicative of the engagement of the person based upon the analyzing of the feature of the person over the predefined time window;
- and
- responsive to computing the value, performing at least one processing function based at least in part upon the value that is indicative of the engagement of the person, the performing of the at least one processing function undertaken by a computer processor.
A computing device, comprising:
- a receiver component that is configured to receive a stream of data from a sensor, the stream of data relating to a person being presented with content;
- a feature extractor component that is configured to extract a feature of the person from the stream of data in real-time;
- an engagement estimator component that is configured to estimate a level of engagement of the person with respect to the content being presented to the person based at least in part upon the feature extracted from the stream of data by the feature extractor component over a predefined window of time, wherein the computing device is configured to perform at least one computing function based upon the level of engagement estimated by the engagement estimator component.
A computer-readable storage medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:
- receiving, from a video camera, a video feed directed at a person while the person is viewing media content;
- extracting, in real-time, a plurality of features of the person from the video feed, the plurality of features comprising face distances identified in the video feed;
- estimating a level of engagement of the person with respect to the media content based upon the plurality of features extracted from the video feed over a predefined window of time, the level of engagement being one of a plurality of predefined levels of engagement;
- and
- updating the media content being viewed by the person based upon the level of engagement of the person estimated with respect to the media content.
Dependencies are inferred from the claim text.