109,664 Hits in 17.0 sec

Indexing Images in High-Dimensional and Dynamic-Weighted Feature Spaces [chapter]

King-Shy Goh, Edward Chang
2002 Visual and Multimedia Information Management  
In this paper. we propose using Tree-structured Vector Quantization (TSVQ) to index high-dimensional data for supporting efficient similarity searches and effective relevance feedback.  ...  When relevance feedback is considered, and thereby features are weighted differently, we show that our approach remains very effective.  ...  To deal with the dimensionality-curse problem and to support dynamic feature weightings, we propose an indexing scheme using clustering and classification methods for supporting approximate similarity  ... 
doi:10.1007/978-0-387-35592-4_10 fatcat:7ekulve5szcsrj77m6ntwpgk6i

Indexing and Integrating Multiple Features for WWW Images

Heng Tao Shen, Xiaofang Zhou, Bin Cui
2006 World wide web (Bussum)  
Both for text and visual feature spaces, the similarity between a point and a local partition's center in individual space is used as the indexing key, where similarity values in different features are  ...  Our LBS method outperforms sequential scan on high dimensional space by an order of magnitude.  ...  others, and visual feature is in standard high dimensional data space.  ... 
doi:10.1007/s11280-006-8560-4 fatcat:abo2roj465gmpjmsp3mgur46fu

A pivot-based index structure for combination of feature vectors

Benjamin Bustos, Daniel Keim, Tobias Schreck
2005 Proceedings of the 2005 ACM symposium on Applied computing - SAC '05  
We present a novel indexing schema that provides efficient nearest-neighbor queries in multimedia databases consisting of objects described by multiple feature vectors.  ...  The benefits of the simultaneous usage of several (statically or dynamically) weighted feature vectors with respect to retrieval effectiveness have been previously demonstrated.  ...  We would also like to thank Stefan Rüger and Peter Howarth for kindly allowing us to use the Corel images database.  ... 
doi:10.1145/1066677.1066945 dblp:conf/sac/BustosKS05 fatcat:wr5ympgfargezmrowr4ybdhoyi

Indexing Text and Visual Features for WWW Images [chapter]

Heng Tao Shen, Xiaofang Zhou, Bin Cui
2005 Lecture Notes in Computer Science  
Both for text and visual feature spaces, the similarity between a point and a local partition's center in individual space is used as the indexing key, where similarity values in different features are  ...  Our LBS method outperforms sequential scan on high dimensional space by an order of magnitude.  ...  others, and visual feature is in standard high dimensional data space.  ... 
doi:10.1007/978-3-540-31849-1_85 fatcat:in2bkrfv4be6jlognxtmzzkdnu

Computing With Contextual Numbers [article]

Vahid Moosavi
2014 arXiv   pre-print
The final products of a trained SOM are a set of ordered (low dimensional) indices and their associated high dimensional weight vectors.  ...  encoder, in which the final low dimensional indices of the trained SOM are pointer to the high dimensional space.  ...  Acknowledgments This work was established at the Singapore-ETH Centre for Global Environmental Sustainability (SEC), Future Cities Laboratory, cofounded by the Singapore National Research Foundation (NRF) and  ... 
arXiv:1408.0889v2 fatcat:7tvwngvm7jahpm3odvleivfpyu

Two-way incremental seriation in the temporal domain with three-dimensional visualization: Making sense of evolving high-dimensional datasets

Peter Wittek
2013 Computational Statistics & Data Analysis  
To achieve a meaningful visualization of high-dimensional data, a compactly supported convolution kernel is introduced, which is similar to filter kernels used in image reconstruction and geostatistics  ...  This filter populates the high-dimensional space with values that interpolate nearby elements, and provides insight into the clustering structure.  ...  The resulting index had 10,074 index terms. We used tf-idf weighting to smooth the space.  ... 
doi:10.1016/j.csda.2013.03.026 fatcat:pc3pupnl4vgydmtzvo7o4gf2e4

Comparison of Multidimensional Data Access Methods for Feature-Based Image Retrieval

Serdar Arslan, Ahmet Sacan, Esra Acar, I. Hakki Toroslu, Adnan Yazici
2010 2010 20th International Conference on Pattern Recognition  
In this paper, we present a rigorous comparison of three different approaches to the image retrieval problem, including cluster-based indexing, distance-based indexing, and multidimensional scaling methods  ...  We show that an optimization of feature contributions to the distance measure can identify irrelevant features and is necessary to obtain the maximum accuracy.  ...  SlimTree SlimTree is a balanced and dynamic paged structure that can be used to index multimedia data efficiently in metric space [3] .  ... 
doi:10.1109/icpr.2010.797 dblp:conf/icpr/ArslanSATY10 fatcat:nn6kveedvbdmldcnv32kupvque

Bridging the Feature Gaps for Retrieval of Multi-Dimensional Images

Jinman Kim, Weidong Cai, Dagan Feng
2009 International Journal of Healthcare Information Systems and Informatics  
We focus on studies where the feature extraction and retrieval were performed using information available in the image's multi-dimensional data spaces.  ...  With the advances in imaging scanners, image complexity and magnitude have also rapidly expanded in both the temporal (time) and spatial (space) dimensions.  ...  (space) such as with computed tomography (CT), and magnetic resonance imaging (MRI); in dynamic temporal sequences (time), i.e., dynamic positron emission tomography (PET), functional MRI, and ultrasound  ... 
doi:10.4018/jhisi.2009010103 fatcat:kytiqt5kyzdqbdsd3zlxagkjce

Fast indexing and searching strategies for feature-based image database systems

Li-Wei Kang
2005 Journal of Electronic Imaging (JEI)  
In practice, different features may have different dynamic ranges and different importances, i.e., different features may have different weighting coefficients.  ...  We propose efficient indexing and searching strategies for feature-based image database systems, in which uncompressed and compressed domain image features are employed.  ...  Chakrabarti and Mehrotra 7 proposed a multidimensional data structure called the hybrid tree for indexing high-dimensional feature spaces.  ... 
doi:10.1117/1.1866148 fatcat:42udk3xjirc7fgl5lods2siygy


King-Shy Goh, Beitao Li, Edward Chang
2002 Proceedings of the tenth ACM international conference on Multimedia - MULTIMEDIA '02  
In addition to its efficiency for conducting similarity searches in very highdimensional spaces, we show that DynDex remains effective when features are weighted dynamically for supporting personalized  ...  We then propose DynDex, an indexing method that deals with both the dynamic and non-metric aspects of the distance function.  ...  It must be able to work in a dynamic and non-metric feature space where only the pairwise-distances between objects are given. 2. It should support high-dimensional indexing efficiently.  ... 
doi:10.1145/641103.641107 fatcat:jmgvmkdiqvanxewq7vbqp7wdcy

Classification-Driven Pathological Neuroimage Retrieval Using Statistical Asymmetry Measures [chapter]

Y. Liu, F. Dellaert, W. E. Rothfus, A. Moore, J. Schneider, T. Kanade
2001 Lecture Notes in Computer Science  
Finally, this selected feature subset is used as indexing features to retrieve medically similar images under a semantic-based image retrieval framework. Quantitative evaluations are provided.  ...  A set of novel image features are computed to quantify the statistical distributions of approximate bilateral asymmetry of normal and pathological human brains.  ...  The image distribution in this two-dimensional feature space can be observed inFigure 4.  ... 
doi:10.1007/3-540-45468-3_79 fatcat:jqnznjdfu5ctnbdp7cr7tjzmu4

Implementation of visual data mining for unsteady blood flow field in an aortic aneurysm

Seiichiro Morizawa, Koji Shimoyama, Shigeru Obayashi, Kenichi Funamoto, Toshiyuki Hayase
2011 Journal of Visualization  
This study was performed to determine the relations between the features of wall shear stress and aneurysm rupture.  ...  For this purpose, visual data mining was performed in unsteady blood flow simulation data for an aortic aneurysm.  ...  The SOM performs nonlinear projection of high-dimensional input dataset onto a two-dimensional map, so that a sequence of close data points in the original high-dimensional space will result in a sequence  ... 
doi:10.1007/s12650-011-0101-2 pmid:22557933 pmcid:PMC3339587 fatcat:z46w6iln2rhuvjrq7yiqwzeowi

3D MARS: immersive virtual reality for content-based image retrieval

M. Nakazato, T.S. Huang
2001 IEEE International Conference on Multimedia and Expo, 2001. ICME 2001.  
3D MARS is an interactive visualization system for Content-Based Image Retrieval (CBIR). In 3D MARS, the user browses and queries images in an immersive 3D Virtual Reality space of CAVE.  ...  The user can examine the result from different view angles by flying-through the space with the joystick. Based on the user's feedback, the system dynamically reorganize its visualization scheme.  ...  Finally, in both text and image retrieval systems, documents are indexed in a high dimensional space. Thus, in order to display documents in a 3D space, the dimensionality has to be reduced.  ... 
doi:10.1109/icme.2001.1237651 dblp:conf/icmcs/NakazatoH01 fatcat:qg2nxixpgnh7jpnt3pmajazsje

Issues in managing image and video data

Shawn D. Newsam, Jelena Tesic, Lei Wang, B. S. Manjunath, Minerva M. Yeung, Rainer W. Lienhart, Chung-Sheng Li
2003 Storage and Retrieval Methods and Applications for Multimedia 2004  
The second half of the paper describes a method for efficiently computing nearest neighbor queries in high-dimensional feature spaces in a relevance feedback framework.  ...  This paper presents an overview of our recent work on managing image and video data. The first half of the paper describes a representation for the semantic spatial layout of video frames.  ...  The nearest neighbor search scheme presented in this paper is one possible solution for dynamic queries in high-dimensional feature spaces. (a) Query image and its semantic layout.  ... 
doi:10.1117/12.538096 dblp:conf/spieSR/NewsamTWM04 fatcat:mzttazcgnzct3efxycqu4j4fju

Effective Image Retrieval Supporting Complex Similarity Queries [chapter]

G.-H. Cha, C.-W. Chung
1998 Visual Database Systems 4 (VDB4)  
In the system, three types of indexes exist: visual indexes for visual information such as color and texture, semantic indexes for non-quantifiable abstract information, and keyword indexes for more abstract  ...  To demonstrate the effectiveness of the retrieval mechanism, a prototype content-based image retrieval system has been designed and implemented.  ...  Therefore, it is appropriate for the dynamic environment where updates are occurred frequently and results in good performance in the queries with high signature weights.  ... 
doi:10.1007/978-0-387-35372-2_22 fatcat:4l5ouw2ho5e3phhpwut6rdmlzy
« Previous Showing results 1 — 15 out of 109,664 results