Event-based image classification and scoring
AssigneeDropbox, Inc.
InventorsBoris Babenko, Nils Peter Welinder
Life of this patent
Estimating the term…
Abstract
Embodiments of the present invention analyze and score each image associated with a group to determine representative image or images for the group. Such analysis can include detecting objects shown in the images, determining the quality of the images, and/or contextually analyzing the images as a group. In some embodiments, each image in a group (e.g., an event) of images can be analyzed by one or more image analysis modules that calculate a score for the image based on a different image characteristic. A composite image score can then be calculated based on the various image characteristic scores to identify the image or images to be used as to represent the group.
Claims20
A method comprising:
- defining an event based on at least one common feature among a plurality of images;
- analyzing each of the plurality of images associated with the event to identify one or more image characteristics within each of the plurality of images, wherein identifying the one or more image characteristics comprises detecting a presence of an object within one or more of the plurality of images;
- computing an image characteristic score for each of the one or more image characteristics, wherein computing the image characteristic score comprises computing an object-detection score based in part on a confidence value associated with the presence of the object within the one or more of the plurality of images;
- computing a composite image score for each image of the plurality of images based on the computed image characteristic score for each of the one or more image characteristics;
- referencing the composite image score for each of the plurality of images to determine, for the event, that at least one image of the plurality of images is representative of the event based on a composite image score computed for the at least one image;
- and
- based on the determination that the at least one image of the plurality of images is representative of the event, causing the at least one image to be displayed for the event at a client device as a thumbnail within a user interface for an online content management service.
A system comprising:
- at least one processor;
- and
- at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
- define an event based on at least one common feature among a plurality of images;
- analyze each of the plurality of images associated with the event to identify one or more image characteristics within each of the plurality of images, wherein identifying the one or more image characteristics comprises detecting a presence of an object within one or more of the plurality of images;
- compute an image characteristic score for each of the one or more image characteristics, wherein computing the image characteristic score comprises computing an object-detection score based in part on a confidence value associated with the presence of the object within the one or more of the plurality of images;
- compute a composite image score for each image of the plurality of images based on the computed image characteristic score for each of the one or more image characteristics;
- reference the composite image score for each of the plurality of images to determine, for the event, that at least one image of the plurality of images is representative of the event based on a composite image score computed for the at least one image;
- and
- based on the determination that the at least one image of the plurality of images is representative of the event, cause the at least one image to be displayed for the event at a client device as a thumbnail within a user interface for an online content management service.
A non-transitory computer readable medium storing instructions thereon that, when executed by at least one processor, cause a computer system to:
- define an event based on at least one common feature among a plurality of images;
- analyze each of the plurality of images associated with the event to identify one or more image characteristics within each of the plurality of images, wherein identifying the one or more image characteristics comprises detecting a presence of an object within one or more of the plurality of images;
- compute an image characteristic score for each of the one or more image characteristics, wherein computing the image characteristic score comprises computing an object-detection score based in part on a confidence value associated with the presence of the object within the one or more of the plurality of images;
- compute a composite image score for each image of the plurality of images based on the computed image characteristic score for each of the one or more image characteristics;
- reference the composite image score for each of the plurality of images to determine, for the event, that at least one image of the plurality of images is representative of the event based on a composite image score computed for the at least one image;
- and
- based on the determination that the at least one image of the plurality of images is representative of the event, cause the at least one image to be displayed for the event at a client device as a thumbnail within a user interface for an online content management service.
Dependencies are inferred from the claim text.