| C. L. Sable and V. Hatzivassiloglou. Text-based approaches for the categorization of images. In Proc. Research and Advanced Technologies for Digital Libraries, Paris, 1999. |
....the information extracted from Yahoo Spain and then it automatically classifies photographs according to their captions. 4. 1 Categorization of Photographs Using Captions Search and categorization of photographs is the center of interest of an important research in the latest years (see, e.g. [3, 15, 16, 18, 22]) Using the information in the pictures (colors, shapes, etc. themselves leads to systems that perform much worse than using the text captions of the images or the text surrounding them [18] For instance, the image categorization system AdEater, designed to detect banners, makes use only of few ....
....is the center of interest of an important research in the latest years (see, e.g. 3, 15, 16, 18, 22] Using the information in the pictures (colors, shapes, etc. themselves leads to systems that perform much worse than using the text captions of the images or the text surrounding them [18]. For instance, the image categorization system AdEater, designed to detect banners, makes use only of few picture features (height, width) but takes advantage of the text surrounding the image (hyperlink text, etc) to perform categorization [10] User Model Search Tex Categorizat ion Yahoo ....
Sable, C.L. and Hatzivassiloglou, V. (1999) Text-Based Approaches for the Categorization of Images. In Proceedings of the European Conference of Digital Libraries (ECDL), published as Research and Advanced Technology for Digital Libraries, lecture Notes in Computer Science 1696, Springer.
....the need for a more advanced approach for the task. It has been shown that for text based image classification pure statistical methods (e. g TF IDF based approaches) can be used for coarse classification into general categories such as for example indoor outdoor images, with great precision [13]. Still, these methods do not perform well when accuracy and user time is of primary importance for a task. A light semantic approach to Image Caption Retrieval is the automatic query expansion for encountering synonymy and polysemy phenomena [11] This usually involves a semantic broadening of ....
C. Sable and V. Hatzivassiloglou. Text-based approaches for the categorization of images. In Proceedings of ECDL, 1999.
....injecting external, linguistic knowledge about what kinds of words are most important for the classification, thus enabling the learning algorithm to zoom in to the most significant input features. For example, in work classifying images on the basis of their captions, Sable and Hatzivassiloglou [14] have established that keeping only the first sentence of image captions and only specific parts of speech significantly improves classification accuracy. Other types of linguistic knowledge have also been found useful for information retrieval tasks (e.g. nominal compounds [6] syntactic ....
C. Sable and V. Hatzivassiloglou. Text-based approaches for the categorization of images. In Proceedings of the 3rd European Conferenceon Researchand AdvancedTechnology for Digital Libraries (ECDL-99), Paris, France, 1999.
No context found.
C. L. Sable and V. Hatzivassiloglou. Text-based approaches for the categorization of images. In Proc. Research and Advanced Technologies for Digital Libraries, Paris, 1999.
No context found.
Sable, C.L., Hatzivassiloglou, V.: Text-based approaches for the categorization of images. In: Proc. Research and Advanced Technologies for Digital Libraries, Paris (1999)
No context found.
C. Sable, V. Hatzivassiloglou, Text-based approaches for the categorization of images, In Proceedings of ECDL, 1999.
No context found.
Sable, C.L., Hatzivassiloglou, V.: Text-based approaches for the categorization of images. In: Proc. Research and Advanced Technologies for Digital Libraries, Paris (1999)
No context found.
C. L. Sable and V. Hatzivassiloglou. Text-based approaches for the categorization of images. In Proc. Research and Advanced Technologies for Digital Libraries, Paris, 1999.
No context found.
C. Sable andV. Hatzivassiloglou, "Text-Based Approaches for the Categorization of Images, Proc. European Conf. Digital Libraries, Lecture Notes in Computer Science, no. 1696, Springer-Verlag, Berlin, 1999, pp. 19--38.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC