Event-based image classification and scoring

Patent metadata

Field	Value
Publication number	US-10140517-B2
Application number	US-201414453284-A
Country	US
Kind code	B2
Filing date	Aug 6, 2014
Priority date	Aug 6, 2014
Publication date	Nov 27, 2018
Grant date	Nov 27, 2018

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.

First claim

What is claimed is:

1. A method comprising:
identifying a plurality of images stored on an online content management service;
analyzing the plurality of images to identify features from the plurality of images;
receiving from a client device, indications of user selections indicating user preferences among the plurality of images;
training a predictive activity model to weight the features from the plurality of images according to the user preferences indicated by the user selections;
determining image characteristic scores associated with the predictive activity model that correspond to the weighted features of the plurality of images;
receiving, at the online content management service, an image from the client device;
generating at least one image characteristic score for at least one feature of the image;
comparing the at least one image characteristic score of the image to the image characteristic scores associated with the predictive activity model to determine that the at least one image characteristic score is similar to one or more of the image characteristic scores; and
based on the determination that the at least one image characteristic score of the image is similar to the one or more of the image characteristic scores, providing, for display on a graphical user interface of the client device, an action recommendation comprising a selectable element to change settings associated with the image.

2. The method of claim 1 , wherein when the at least one image characteristic score of the image is below a threshold that indicates that the image is off-center, blurry, or that the subject's eyes are closed, the action recommendation comprises a recommendation to retake the image or perform a corrective action.

3. The method of claim 1 , wherein when the at least one image characteristic score of the image is below a threshold that indicates that the image is dark or blurry, the action recommendation comprises the selectable element to activate the client device's flash.

4. The method of claim 1 , wherein the plurality of images comprises a plurality of popular images based on detected activity associated with the plurality of popular images and the action recommendation comprises a recommendation to share the image.

5. The method of claim 1 , wherein when the at least one characteristic score is below a threshold that indicates that the age is off-center, blurry, or that the subject's eyes are closed, the action recommendation comprises a recommendation to delete the image.

6. The method of claim 1 , wherein receiving the indications of user selections indicating user preferences among the plurality of images comprises receiving a ranking by a user of the plurality of images.

7. The method of claim 1 , wherein receiving the indications of user selections indicating user preferences among the plurality of images comprises receiving indications of user preferences for the identified features from the plurality of images.

8. The method of claim 1 , wherein:
generating the at least one image characteristic score for at least one feature of the image comprises generating a facial-detection score based in part on a confidence value associated with a presence of a person's face within the image; and
the action recommendation comprises a recommendation to share the image with the person whose face is identified in the image.

9. A system comprising:
at least one processor; and
at least one on-transitory computer readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
identify a plurality of images stored on an online content management service;
analyze the plurality of images to identify features from the plurality of images;
receive, from a client device, indications of user selections indicating user preferences among the plurality of images;
train a predictive activity model to weight the features from the plurality of images according to the user preferences indicated by the user selections;
determine image characteristic scores associated with the predictive activity model that correspond to the weighted features of the plurality of images;
receive, at the online content management service, an image from the client device;
generate at least one image characteristic score for at least one feature of the image;
compare the at least one image characteristic score of the image to the image characteristic scores associated with the predictive activity model to determine that the at least one image characteristic score is similar to one or more of the image characteristic scores; and
based on the determination that the at least one image characteristic score of the image is similar to the one or more of the image characteristic scores, provide, for display on a graphical user interface of the client device, an action recommendation comprising a selectable element to change settings associated with the image.

10. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to provide the action recommendation comprising the selectable element to change settings associated with the image by, when the at least one image characteristic score of the image is below a threshold that indicates that the image is off-center, blurry, or that the subject's eyes are closed, providing the action recommendation comprising a recommendation to retake the image or perform a corrective action.

11. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to provide the action recommendation comprising the selectable element to change settings associated with the image by, when the at least one image characteristic score of the image is below a threshold that indicates that the image is dark or blurry, providing the action recommendation comprising the selectable element to activate the client device's flash.

12. The system of claim 9 , further comprising instruction that, when executed by the at least one processor, cause the system to:
identify the plurality of images stored on the online content management service by identifying a plurality of popular images based on detected activity associated with the plurality of popular images; and
provide the action recommendation comprising the selectable element to change settings associated with the image by providing the action recommendation comprising a recommendation to share the image.

13. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to provide the action recommendation comprising the selectable element to change settings associated with the image by, when the at least one characteristic score is below a threshold that indicates that the image is off-center, blurry, or that the subject's eyes are closed, providing the action recommendation comprising the selectable element to deactivate the client device's flash.

14. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to receive the indications of user selections indicating user preferences among the plurality of images by receiving indications of user preferences for the identified features from the plurality of images.

15. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to receive the indications of user selections indicating user preferences among the plurality of images by receiving a ranking by a user of th

Assignees

• Dropbox Inc

Inventors

• Babenko Boris
• Welinder Nils Peter

Classifications

• G06K9/00744 Primary

  Physics · mapped topic

• G06V20/46 Primary

  Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames · CPC title