Generation of Annotation Tags Based on Multimodal Metadata and Structured Semantic Descriptors
1 Assignment
0 Petitions
Accused Products
Abstract
In one embodiment, a method of generating annotation tags (28) for a digital image (22) includes maintaining a library (16) of human-meaningful words or phrases organized as category entries (72) according to a number of defined image description categories (70), and receiving context metadata (20) associated with the capture of a given digital image (22). The method further includes selecting particular category entries (72-1, 72-2) as vocabulary metadata (24) for the digital image (22) by mapping the context metadata (20) into the library (16), and generating annotation tags (28) for the digital image (22) by logically combining the vocabulary metadata (24) according to a defined set of deductive logic rules (30) that are predicated on the defined image description categories (70). In another embodiment, a processing apparatus (12), such as a digital processor (18, 26) and supporting memory (14), etc., is configured to carry out the above method, or to carry out variations of the above method.
52 Citations
32 Claims
-
1-16. -16. (canceled)
-
17. A method of generating annotation tags for a digital image characterized by:
-
maintaining a library of human-meaningful words or phrases organized as category entries according to a number of defined image description categories; receiving context metadata for the digital image that is associated with its capture; selecting particular category entries as vocabulary metadata for the digital image by mapping the context metadata into the library; generating annotation tags for the digital image by logically combining the vocabulary metadata according to a defined set of deductive logic rules that are predicated on the defined image description categories, said defined set of deductive logic rules comprising a set of conditional tests predicated on variable entries for one or more of the defined image description categories, such that said generating annotation tags for the digital image comprises substituting the category entries making up the vocabulary metadata as values for corresponding ones of the variable entries, and then logically evaluating the set of conditional tests according to the substituted category entries; determining confidence or fuzzy logic weighting values from the context metadata, said confidence or fuzzy logic weighting values numerically expressing the extents to which given ones of the substituted category entries apply to the digital image; and ranking the generated annotation tags from most likely to least likely, in terms of applicability to the digital image, according to the confidence or fuzzy logic weighting values, or incorporating the confidence or fuzzy logic weighting values into said logical evaluation of the set of conditional tests, whereby the truthfulness of each conditional evaluation depends on the confidence or fuzzy logic weighting values, such that the generated annotation tags correspond to higher-weighted category entries or combinations thereof. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24)
-
-
25. A processing apparatus for generating annotation tags for a digital image, said processing apparatus comprising an electronic circuit that comprises:
-
a memory for maintaining a library of human-meaningful words or phrases organized as category entries according to a number of defined image description categories; an abstraction processor configured to receive context metadata for the digital image that is associated with its capture and select particular category entries as vocabulary metadata for the digital image by mapping the context metadata into the library; and a rules processor configured to generate annotation tags for the digital image by logically combining the vocabulary metadata according to a defined set of deductive logic rules that are predicated on the defined image description categories; wherein said defined set of deductive logic rules comprises a set of conditional tests predicated on variable entries for one or more of the defined image description categories; wherein said rules processor is configured to generate annotation tags for the digital image by substituting the category entries making up the vocabulary metadata as values for corresponding ones of the variable entries, and then logically evaluating the set of conditional tests according to the substituted category entries; and wherein said processing apparatus is further configured to; determine confidence or fuzzy logic weighting values from the context metadata, said confidence or fuzzy logic weighting values numerically expressing the extents to which given ones of the substituted category entries apply to the digital image; and rank the generated annotation tags from most likely to least likely, in terms of applicability to the digital image, according to the confidence or fuzzy logic weighting values, or incorporate the confidence or fuzzy logic weighting values into said logical evaluation of the set of conditional tests, whereby the truthfulness of each conditional evaluation depends on the confidence or fuzzy logic weighting values, such that the generated annotation tags correspond to higher-weighted category entries or combinations thereof. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32)
-
Specification