Object-level image editing
Systems and methods for editing digital images using information about objects in those images are described. For example, the information about objects comprises depth ordering information and/or information about the class each object is a member of. Examples of classes include sky, building, aeroplane, grass and person. This object-level information is used to provide new and/or improved editing functions such as cut and paste, filling-in image regions using tiles or patchworks, digital tapestry, alpha matte generation, super resolution, auto cropping, auto color balance, object selection, depth of field manipulation, and object replacement. In addition improvements to user interfaces for image editing systems are described which use object-level information.
1. A method of editing a source digital image comprising:
accessing automatically generated object-level information about the source digital image, the object-level information including:
a segmentation of the source digital image into one or more objects, and
for each of the one or more objects, one or more details of a class that the object is a member of;
receiving a user input to select one of the objects in the source digital image based on the object-level information;
generating replacement material based on the object-level information for the selected object, generating including obtaining replacement material of a first size from a source; and
automatically processing the source digital image based on the selected object and the object-level information, including replacing the selected object with the replacement material to form an output digital image, automatically processing including:
automatically determining, independent of any user size selection, to scale the replacement material based at least in part on relative sizes of two or more of the replacement material, the selected object, and at least another object of the one or more objects; and
automatically scaling the replacement material to match the user-selected object in the source digital image using at least a portion of the automatically generated object-level information, automatically scaling including changing the replacement material from the first size to a second size, the replacement material of the second size being absent from the source.
2. A method as claimed in claim 1 wherein the object-level information includes at least one of: information about a depth ordering of at least some of the one or more objects, one or more object attributes, or information about a position of a part of at least one of the objects.
3. A method as claimed in claim 1 further comprising:
combining at least part of the source digital image with one or more reference digital images to form the output digital image; and
accessing automatically generated object-level information about the one or more reference digital images including:
segmentation of the one or more reference digital images into one or more objects and,
for each of the one or more objects, details of a class that the object is a member of.
4. A method as claimed in claim 1 carried out for a plurality of source digital images to create a single output image and wherein the single output image is formed automatically based on object-level information about each of the plurality of source digital images.
5. A method as claimed in claim 4 further comprising selecting one or more parts of the plurality of source digital images for inclusion in the single output image on the basis of the object-level information.
6. A method as claimed in claim 4 further comprising selecting one or more locations in the single output image to position one or more parts of the plurality of source digital images based on the object-level information.
7. A method as claimed in claim 1 wherein the source digital image has a lower resolution compared to the output digital image.
8. A method as claimed in claim 7 wherein automatically processing the source digital image further comprises accessing a database of images having higher resolution than the source digital image, and selecting one or more images from the database to use in forming the output digital image, based on the object-level information.
9. A method as claimed in claim 1 wherein the output digital image is cropped from the source digital image and wherein automatically processing the source digital image further comprises automatically cropping the source digital image based on the object-level information.
10. A method as claimed in claim 9 further comprising automatically cropping the source digital image based on at least one of a template, a rule, or one or more selected criteria.
11. A method as claimed in claim 1 wherein automatically processing the source digital image further comprises adjusting the color balance of the source digital image on the basis of the object-level information.
12. A method as claimed in claim 1 further comprising presenting an indication to a user via a user interface as to which object in the source digital image is selected as a result of the single user input, the selection having been made by the user with a single action.
13. A method as claimed in claim 12 further comprising arranging the user interface such that editing options presented to a user via the interface are selected on the basis of the object-level information associated with the selected object.
14. A method as claimed in claim 12 further comprising modifying at least one of a color, a focus or a depth order of the selected object.
15. A method as claimed in claim 1 wherein automatically processing the source digital image further comprises replacing an object or part of an object in the source digital image with an image of a selected object or object part from a database and wherein selection of the object or object part is made based on the object-level information.
16. An apparatus comprising:
an interface to:
access automatically generated object-level information about a source digital image, the object-level information including a segmentation of the source digital image into one or more objects and, for each of the one or more objects, one or more details of a class that the object is a member of, and
receive a single input selecting one of the objects in the source digital image based on the object-level information; and
at least one processor to:
determine a replacement object of a related class to the class of the selected object based on the object-level information for the selected object, determining the replacement object including obtaining the replacement object of a first size from a source; and
automatically process the source digital image based on the selected object and the object-level information, including replacing the selected object with the replacement object to form an output digital image, automatically processing including:
automatically determining, independent of any user size selection, to scale the replacement object based at least in part on relative sizes of two or more of the replacement object, the selected object, and at least another object of the one or more objects; and
automatically scaling the replacement object to match the user-selected object in the source digital image using at least a portion of the automatically generated object-level information, automatically scaling including changing the replacement object from the first size to a second size, the replacement material of the second size being absent from the source.
17. An apparatus as claimed in claim 16 wherein the object-level information includes at least one of information about a depth ordering of at least some of the objects, one or more object attributes, and information about a position of a part of at least one of the objects.
18. An apparatus as claimed in claim 16 wherein the processor further operates to combine at least a part of the source digital image with one or more reference digital images to form the output digital image and wherein the interface further operates to access automatically generated object-level information about the one or more reference digital images, the information including a segmentation of the one or more reference digital images into one or more objects and, for each of the objects, one or more details of a class that the object is a member of.
19. One or more computer readable storage media, the one or more computer readable storage media being hardware, storing instructions that, when executed by a processor, enable the processor to perform actions comprising:
accessing automatically generated object-level information associated with a source digital image, the object-level information including a segmentation of the source digital image into one or more objects and, for each of the one or more objects, one or more details of a class that the object is a member of; and
automatically processing the source digital image based on the object-level information to form an output digital image, including:
automatically cropping the source digital image based on a single input to select one of the objects in the source digital image;
determining a replacement object of a related class to the class of at least one object that is adjacent to the selected object in the source digital image based on the object-level information, determining including obtaining the replacement object of a first size from a source;
automatically determining, independent of any user size selection, to scale the replacement object based at least in part on relative sizes of two or more of the replacement object, the selected object, and at least another object of the one or more objects;
automatically scaling the replacement object to match the selected object using at least a portion of the automatically generated object-level information, automatically scaling including changing the replacement object from the first size to a second size, the replacement material of the second size being absent from the source; and
replacing the selected object with the replacement object to form the output digital image.
20. The one or more computer readable storage media of claim 19 , wherein the object-level information includes at least one of: information about a depth ordering of at least some of the one or more objects, one or more object attributes, and information about a position of one or more object parts.