|
Accession number;07A0176782
|
| Title;An Analysis of "Visualness" of Word Concepts |
| Author;
YANAI KEIJI
(Univ. of Electro-Communications)
BARNARD KOBUS
(Univ. Arizona, USA)
|
Journal Title;IPSJ Transactions on Database
|
Journal Code:Z0778A
|
ISSN:0387-5806
|
|
VOL.48;NO.SIG1(CVIM17);PAGE.88-97(2007)
|
| Figure&Table&Reference;FIG.6, TBL.6, REF.20 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;We propose a new method to measure "visualness" of word concepts, that is, what extent concepts have visual characteristics. To know which concept has visually discriminative power is important for image annotation, especially automatic image annotation by image recognition system, since not all concepts are related to visual contents. Our method performs probabilistic region selection for images which are labeled as concept "X" or "non-X", and computes an entropy measure which represents "visualness" of concepts. In the experiments, we collected forty-five thousand images from the World-Wide Web using the Google Image Search for 150 concepts. We examined which concepts were suitable for annotation of image contents. (author abst.) |
|
|
|
Related Articles;
|
