Method and apparatus for representing and searching for an object in an image
A method representing an object appearing in still or video image for use in searching, wherein the object appears in the image with a first two-dimensional outline, by processing signals corresponding to the image, comprises deriving a view descriptor of the first outline of the object and deriving at least one additional view descriptor of the outline of the object in a different view, and associating the two or more view descriptors to form an object descriptor.
1. A method of representing an object appearing in an image using one or more processors, the method comprising:
deriving, using a processor, a plurality of view descriptors of the object, each view descriptor corresponding to a different view of the object;
indicating for each view descriptor whether or not the respective view corresponds to a view of the object appearing in the image by associating, using a processor, a visibility flag with each view descriptor, the visibility flag indicating whether or not the view corresponds to a view of the object appearing in the image,
wherein at least one view descriptor comprises a representation of the texture of the object in the respective view, wherein each view is a view of the object from a respective angle of view, the plurality of view descriptors including at least one descriptor of the view of the object in the image and at least one descriptor of a view of the object from an angle of view different from the angle of view of the object in the image.
2. A method as claimed in claim 1 comprising associating the view descriptors to form an object descriptor.
3. A method as claimed in claim 1 which is for indexing the image or collection of images for searching purposes.
4. A method of searching for a query object in stored images and groups of images, the method comprising inputting a query, matching the query with representations of the stored images derived using a method as claimed in claim 1 , and retrieving the closest matches, wherein the query input includes indicating whether or not the search should be limited to images or groups of images which show the query object in the same view as the query view.
5. An apparatus an image database storing still images or groups of images;
a descriptor database storing descriptors for objects appearing in the images stored in the image database; and
a processor configured to perform the method as claimed in claim 1 .
6. A computer-readable storage medium storing computer-executable process steps, for implementing a method as claimed in claim 1 .
7. A computer system comprising:
an image database storing still image or groups of images;
a descriptor database storing descriptors for objects appearing in the images stored in the image database; and
a control unit programmed to perform the method as claimed in claim 1 .
8. A method as claimed in claim 1 , wherein said visibility flag is a first value if the view corresponds to a view of the object appearing in the image and a second value if the view corresponds to a view of the object not appearing in the image.
9. A method of representing an object appearing in a group of images using one or more processors, the method comprising:
deriving, using a processor, a plurality of view descriptors of the object, each view descriptor corresponding to a different view of the object;
indicating for each view descriptor whether or not the respective view corresponds to a view of the object appearing in the group of images, by associating, using a processor, a visibility flag with each view descriptor, the visibility flag indicating whether or not the view corresponding to a view of the object appearing in the group of images,
wherein at least one view descriptor comprises a representation of the texture of the object in the respective view, wherein each view is a view of the object from a respective angle of view, the plurality of view descriptors including at least one descriptor of the view of the object in the image and at least one descriptor of a view of the object from an angle of view different from the angle of view of the object in the image.
10. A method as claimed in claim 9 comprising indicating, for each image in the group of images, for each view descriptor whether or not the respective view corresponds to a view of the object appearing in said image.
11. A method as claimed in claim 9 wherein the group of images is a sequence of images.
12. A method as claimed in claim 11 wherein the sequence of images is from a video or film.
13. A method as claimed in claim 9 , wherein said visibility flag is a first value if the view corresponds to a view of the object appearing in the group of images and a second value if the view corresponds to a view of the object not appearing in the group of images.