Method and system for multi-tier image matching in a mixed media environment
First Claim
Patent Images
1. A computer-implemented method of image matching, the method comprising:
- receiving, with one or more processors, a first image of at least part of a document in a first media type;
identifying, with the one or more processors, a culling feature in the first image by comparing the first image to a library of images, the library of images comprising at least one image that includes the culling feature;
identifying, with the one or more processors, a location of the culling feature in the first image;
selecting, with the one or more processors, a set of documents that include the culling feature and the location from a collection of stored documents indexed by the culling feature;
receiving, with the one or more processors, a second image of at least part of the document in the first media type that is distinct from the first image;
identifying, with the one or more processors, horizontal and vertical bounding boxes in the second image that include horizontally and vertically adjacent pairs of words;
determining, with the one or more processors, a subset of documents from the selected set of documents by identifying a number of overlaps of the horizontal and vertical bounding boxes and locations of centers of the overlaps in the selected set of documents, wherein each document in the subset includes at least one of the horizontally and vertically adjacent pairs of words; and
ranking, with the one or more processors, the subset of documents based on the number of overlaps of the horizontal and vertical bounding boxes and the locations of the centers of the overlaps.
1 Assignment
0 Petitions
Accused Products
Abstract
A Mixed Media Reality (MMR) system and associated techniques are disclosed. The MMR system provides mechanisms for forming a mixed media document that includes media of at least two types (e.g., printed paper as a first medium and digital content and/or web link as a second medium). In one particular embodiment, the MMR system provides for multi-tier image matching.
-
Citations
23 Claims
-
1. A computer-implemented method of image matching, the method comprising:
-
receiving, with one or more processors, a first image of at least part of a document in a first media type; identifying, with the one or more processors, a culling feature in the first image by comparing the first image to a library of images, the library of images comprising at least one image that includes the culling feature; identifying, with the one or more processors, a location of the culling feature in the first image; selecting, with the one or more processors, a set of documents that include the culling feature and the location from a collection of stored documents indexed by the culling feature; receiving, with the one or more processors, a second image of at least part of the document in the first media type that is distinct from the first image; identifying, with the one or more processors, horizontal and vertical bounding boxes in the second image that include horizontally and vertically adjacent pairs of words; determining, with the one or more processors, a subset of documents from the selected set of documents by identifying a number of overlaps of the horizontal and vertical bounding boxes and locations of centers of the overlaps in the selected set of documents, wherein each document in the subset includes at least one of the horizontally and vertically adjacent pairs of words; and ranking, with the one or more processors, the subset of documents based on the number of overlaps of the horizontal and vertical bounding boxes and the locations of the centers of the overlaps. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A system for image matching, comprising:
-
one or more processors; a memory coupled to the one or more processors; and a database including stored instructions executable by the one or more processors, the database operable to receive a first image of at least part of a document in a first media type, to identify a culling feature in the first image by comparing the first image to a library of images, the library of images comprising at least one image that includes the culling feature, to identify a location of the culling feature in the first image, to select a set of documents that include the culling feature and the location from a collection of documents indexed by the culling feature and stored within the database, to receive a second image of at least part of the document in the first media type that is distinct from the first image, to identify horizontal and vertical bounding boxes in the second image that include horizontally and vertically adjacent pairs of words, to determine a subset of documents from the selected set of documents by identifying a number of overlaps of the horizontal and vertical bounding boxes and locations of centers of the overlaps in the selected set of documents, wherein each document in the subset includes at least one of the horizontally and vertically adjacent pairs of words and to rank the subset of documents based on the number of overlaps of the horizontal and vertical bounding boxes and the locations of the centers of the overlaps. - View Dependent Claims (9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A computer-implemented method of providing interaction between a first media type with a second media type, comprising:
-
receiving, with one or more processors, a first image of at least part of a document in the first media type; identifying, with the one or more processors, a culling feature in the first image by comparing the first image to a library of images, the library of images comprising at least one image that includes the culling feature; identifying, with the one or more processors, a location of the culling feature in the first image; selecting, with the one or more processors, a set of documents that include the culling feature and the location from a document database indexed by the culling feature; receiving, with the one or more processors, a second image of at least part of the document in the first media type that is distinct from the first image; identifying, with the one or more processors, horizontal and vertical bounding boxes in the second image that include horizontally and vertically adjacent pairs of words; determining, with the one or more processors, a subset of documents from the selected set of documents by identifying a number of overlaps of the horizontal and vertical bounding boxes and locations of centers of the overlaps in the selected set of documents, wherein each document in the subset includes at least one of the horizontally and vertically adjacent pairs of words; ranking, with the one or more processors, the subset of documents based on the number of overlaps of the horizontal and vertical bounding boxes and the locations of the centers of the overlaps; and providing, with the one or more processors, a second media type associated with a highest ranked document from the subset of documents responsive to determining the subset of documents. - View Dependent Claims (19, 20, 21, 22)
-
-
23. A computer program product comprising a non-transitory computer readable medium including a computer readable program, wherein the computer readable program when executed on a computer causes the computer to:
-
receive a first image of at least part of a document in a first media type; identify a culling feature in the first image by comparing the first image to a library of images, the library of images comprising at least one image that includes the culling feature; identify a location of the culling feature in the first image; select a set of documents that includes the culling feature and the location from a collection of documents indexed by the culling feature and stored within a database that includes the culling feature and the location; receive a second image of at least part of the document in the first media type that is distinct from the captured first image; identify horizontal and vertical bounding boxes in the second image that include horizontally and vertically adjacent pairs of words; determine a subset of documents from the selected set of documents by identifying a number of overlaps of the horizontal and vertical bounding boxes and locations of centers of the overlaps in the selected set of documents, wherein each document in the subset includes at least one of the horizontally and vertically adjacent pairs of words; and rank the subset of documents based on the number of overlaps of the horizontal and vertical bounding boxes and the locations of the centers of the overlaps.
-
Specification