Method and apparatus for identifying a representative area of an image
AssigneeAdobe Systems Incorporated
InventorsMohit Gupta, Durga Ganesh Grandhi
Life of this patent
Estimating the term…
Abstract
A computer implemented method for generating a representative thumbnail for an image. The method comprises determining a representative area of an image, the determining comprising determining an absence of faces in the image; dividing the image into one or more zones; and selecting a zone with maximum edge strength as the representative area; and generating a thumbnail by cropping the image to the representative area.
Claims20
A computer implemented method comprising:
- determining a representative area of an image, the determining comprising:
- determining an absence of faces in the image;
- dividing the image into one or more zones;
- and
- selecting a zone with maximum edge strength as the representative area;
- and
- generating a thumbnail by cropping the image to the representative area.
An apparatus comprising a computer having one or more processors for executing instructions comprising:
- a face detection module configured to:
- detect that a scaled image includes one or more faces;
- determine a priority for each of the one or more faces;
- and
- an image processing module configured to:
- select a representative area, wherein the representative area includes the one or more faces, when the priority of each of the one or more faces is greater than a predefined threshold;
- and
- generate a thumbnail by cropping the image to the representative area.
A non-transitory computer readable medium for storing computer instructions that, when executed by at least one processor causes the at least one processor to perform a method comprising:
- determining a representative area of an image, the determining comprising:
- determining an absence of faces in the image;
- dividing the image into one or more zones;
- and
- selecting a zone with maximum edge strength as the representative area;
- and
- generating a thumbnail by cropping the image to the representative area.
Dependencies are inferred from the claim text.