J.-Y. Chen, C. A. Bouman, and J. Dalton, "Active browsing using similarity pyramids," in Proc. 36th Asilomar Conf. Signals, Systems, Computers, Pacific Grove, CA, Nov. 4, 1998.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....temporal order or clustered based on content similarity to aid browsing. Browsing tools for large image databases have been presented using clustering techniques. The use of hierarchical clustering technique for fast image search and non linear browsing has been shown [9] An active browsing tool [8] that uses hierarchical clustering and relevance feedback has also been presented recently. The drawbacks of the browsing techniques that use clustering techniques are: 1) they require computationally expensive clustering algorithms, 2) addition or deletion of content from the database may ....

J-Y. Chen, C.A. Bouman and J. Dalton, "Active Browsing using Similarity Pyramids", In Storage and Retrieval for Image and Video Databases VII, Vol. 3656, SPIE Proceedings, Jan 1998.

.... is generated as a weighted linear combination of the original feature vector and the feature vectors of the images that were labeled as relevant or irrelevant by the user [12, 6] Other approaches include using keywords [8] creating a probabilistic user model [7] modifying the distance measure [4], reorganizing the retrieval results [5] feature density estimation [10] and iterative weight updating [13, 11] Retrieval algorithms depend on features directly computed from images. We want to use only the information fed back by the user instead of using artificial keywords or heuristic ....

J.-Y. Chen, C. Bouman, and J. Dalton. Active browsing using similarity pyramids. In Proceedings of the VII IS&T/SPIE Conference on Storage and Retrieval for Image and Video Databases, volume 3656, pages 144--154, San Jose, CA,, January 1999.

.... is generated as a weighted linear combination of the original feature vector and the feature vectors of the images that were labeled as relevant or irrelevant by the user [10, 4] Other approaches include using keywords [6] creating a probabilistic user model [5] modifying the distance measure [3], feature density estimation [8] and iterative weight updating [11, 9] Retrieval algorithms depend on features directly computed from images. We want to use only the information feedbacked by the user instead of using artificial keywords or heuristic assumptions. We also cannot assume anything ....

J.-Y. Chen, C. Bouman, and J. Dalton. Active browsing using similarity pyramids. In Proceedings of the VII IS&T/SPIE Conference on Storage and Retrieval for Image and Video Databases, volume 3656, pages 144--154, San Jose, CA,, January 1999.

....measure for the pyramid organization which we call dispersion, and use this measure to evaluate various approaches to pyramid design. Finally, we note that the image browsing environments described in this research can be adapted for specific search tasks through the use of relevance feedback [22], 23] II. GENERAL APPROACH AND NOTATION Let the images in the database be indexed by where is the complete set of images. Each image will have an associated feature vector which contains the relevant information required for measuring the similarity between images. Furthermore, we will assume ....

J.-Y. Chen, C. A. Bouman, and J. Dalton, "Active browsing using similarity pyramids," in Proc. 36th Asilomar Conf. Signals, Systems, Computers, Pacific Grove, CA, Nov. 4, 1998.

No context found.

J-Y. Chen, C.A. Bouman and J. Dalton, "Active Browsing using Similarity Pyramids", In Storage and Retrieval for Image and Video Databases VII, Vol. 3656, SPIE Proceedings, Jan 1998.

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