Method and system for automatically cropping images
AssigneeGoogle Inc.
InventorsClint Chester N. Chua, Sally Goldman, Jonathan Albert Wray
Life of this patent
Estimating the term…
Abstract
An image cropping technique includes detecting, at a computer system including one or more processors, faces within an image and evaluating, at the computer system, the faces to determine if any of the faces correspond to people within a user's online social network. The technique additionally includes defining, at the computer system, a first crop region including a subset of the faces when at least one face within the subset of the faces corresponds to a first person in the user's online social network and cropping, at the computer system, the image based on the first crop region.
Claims16
A computer-implemented method comprising:
- detecting, at a computer system including one or more processors, faces within an image;
- evaluating, at the computer system, the faces to determine if any of the faces correspond to people within a user's online social network;
- determining, at the computer system, a degree of affiliation between the user and each person in the subset of the faces from the user's online social network;
- defining, at the computer system, a first crop region including a subset of the faces when at least one face within the subset of the faces corresponds to a first person in the user's online social network based on the degree of affiliation;
- reducing, at the computer system, saliency values associated with regions of the image not included in the first crop region;
- and
- cropping, at the computer system, the image based on the first crop region.
A computer-implemented method comprising:
- detecting, at a computer system including one or more processors, faces within an image;
- evaluating, at the computer system, the faces to determine if any of the faces correspond to people within a user's online social network;
- determining, at the computer system, a first degree of affiliation between the user and the first person;
- defining, at the computer system, a first crop region including a subset of the faces when at least one face within the subset of the faces corresponds to a first person in the user's online social network, wherein the crop region is defined to include the at least one face when the first degree of affiliation is within a predetermined threshold;
- and
- cropping, at the computer system, the image based on the first crop region;
- detecting, at the computer system, a second face within the image;
- evaluating, at the computer system, the second face;
- identifying, at the computer system, a second person from the user's online social network corresponding to the second face based on the evaluating;
- determining, at the computer system, a second degree of affiliation between the user and the second person;
- and
- defining, at the computer system, the first crop region to include the at least one face and exclude the second face when the first degree of affiliation is greater than the second degree of affiliation.
A system for automatic cropping of images comprising:
- a face detector that detects faces within an image and determines identities of the faces relative to people within a user's online social network;
- a saliency analyzer that determines a degree of affiliation between the user and each person in the subset of the faces from the user's online social network and that reduces saliency values associated with regions of the image not included with the at least one face;
- a search system that determines a first crop region including a subset of the faces when at least one face within the subset of the faces corresponds to a first person in the user's online social network based on the degree of affiliation;
- and
- a cropping system that crops the image based on the first crop region.
Dependencies are inferred from the claim text.