Filters








119,588 Hits in 4.3 sec

Fast Approximate Point Set Matching for Information Retrieval [chapter]

Raphaël Clifford, Benjamin Sach
2007 Lecture Notes in Computer Science  
We investigate randomised algorithms for subset matching with spatial point sets-given two sets of d-dimensional points: a data set T consisting of n points and a pattern P consisting of m points, find  ...  the largest match for a subset of the pattern in the data set.  ...  The authors would like to thank Manolis Christodoulakis for the original implementation of the MSM-FT algorithm and the EPSRC for the funding of the second author.  ... 
doi:10.1007/978-3-540-69507-3_17 fatcat:l6bx3oqwyvfpdcybekz4qvywb4

Minimizing the Number of Keypoint Matching Queries for Object Retrieval

Johannes Niedermayer, Peer Kröger
2015 Procedings of the British Machine Vision Conference 2015  
To increase the efficiency of feature matching, we rely on fast state-of-the-art (approximate) indexing techniques optimized for high-dimensional data, for example Locally Optimized Product Quantization  ...  However, with the vast amount of features that have to be matched during recognition (up to a few thousand), even very fast kNN indexing techniques that can provide approximate query results in under ten  ...  To increase the efficiency of feature matching, we rely on fast state-of-the-art (approximate) indexing techniques optimized for high-dimensional data, for example Locally Optimized Product Quantization  ... 
doi:10.5244/c.29.48 dblp:conf/bmvc/NiedermayerK15 fatcat:rsgrr2kahbftdi6ineor6k6wza

Content-based object retrieval using maximum curvature points in contour images

Azhar Quddus, Faouzi A. Cheikh, Moncef Gabbouj, Minerva M. Yeung, Boon-Lock Yeo, Charles A. Bouman
1999 Storage and Retrieval for Media Databases 2000  
In this paper we present a technique for shape similarity estimation and apply it to content-based indexing and retrieval over large image databases.  ...  This technique is based on wavelet decomposition and uses polygonal approximation.  ...  Authors also acknowledge Tampere Graduate School in Information Science and Engineering (TISE) and Tampere International Center for Signal Processing (TICSP) for providing the financial support for this  ... 
doi:10.1117/12.373540 dblp:conf/spieSR/QuddusCG00 fatcat:clfbq6ij5vcvzdaehkxrpvzady

Commercial Break Detection and Content Based Video Retrieval [chapter]

N. Venkatesh, M. Girish Chandra
2010 Lecture Notes in Electrical Engineering  
Detected splash screen of a program at the commercial boundaries is used for automatic annotation of recorded video which helps in fast content based video retrieval.  ...  This chapter presents a novel approach for automatic annotation and content based video retrieval by making use of the features extracted during the process of detecting commercial boundaries in a recorded  ...  The correlation matching technique is simple and computationally fast and hence it has been used to find the best match between the query image and the set of images in the database.  ... 
doi:10.1007/978-90-481-9419-3_29 fatcat:nhrwrgcthjc5ngg3cmw63qti6a

Efficient Query Specific DTW Distance for Document Retrieval with Unlimited Vocabulary

Gattigorla Nagendar, Viresh Ranjan, Gaurav Harit, C. Jawahar
2018 Journal of Imaging  
In (DQC), the classifiers are trained for a set of frequent queries and seamlessly extended for the rare and arbitrary queries.  ...  Unlike previous approaches, the proposed query specific DTW distance uses both the class mean vectors and the query information for computing the global principal alignments for the query.  ...  Acknowledgments: This work was supported from the grant received for the IMPRINT project titled "Information access from document images of Indian languages," from MHRD, Government of India.  ... 
doi:10.3390/jimaging4020037 fatcat:c6ckibqanbaidgcrqrs7fot6cu

Similarity-Invariant Sketch-Based Image Retrieval in Large Databases [chapter]

Sarthak Parui, Anurag Mittal
2014 Lecture Notes in Computer Science  
Chain similarities are computed by a fast Dynamic Programming-based approximate substring matching algorithm, which enables partial matching of chains.  ...  Finally, hierarchical k-medoids based indexing is used for very fast retrieval in a few seconds on databases with millions of images.  ...  Second, an efficient DP-based approximate substring matching algorithm is proposed for fast matching of such chains between a sketch and an image or between two images.  ... 
doi:10.1007/978-3-319-10599-4_26 fatcat:lp4xho4srzekzblb457wnr7djy

SIFT-Based Low-Quality Fingerprint LSH Retrieval and Recognition Method

Yunfei Zhong, Xiaoqi Peng
2015 International Journal of Signal Processing, Image Processing and Pattern Recognition  
and matching mechanism for the automated recognition of the large-scale fingerprint database, with strong practicality.  ...  Experimental results show that the proposed method has strong penetration, high retrieval efficiency, good recognition accuracy, and strong robustness, thereby providing a fast and efficient retrieval  ...  Then, the method uses the scale grouping principle to perform reduction recognition, thus providing a fast and efficient retrieval and matching mechanism for the automatic recognition of large-scale fingerprint  ... 
doi:10.14257/ijsip.2015.8.8.28 fatcat:bjlzdltxhfebpauzfsma5ckmay

RETRIEVAL FROM SHAPE DATABASES USING CHANCE PROBABILITY FUNCTIONS AND FIXED CORRESPONDENCE

BOAZ J. SUPER
2006 International journal of pattern recognition and artificial intelligence  
In the past, the matchers used for the oneto-many problem of shape retrieval were often designed for the problem of matching an isolated pair of shapes.  ...  (4) It has a fast runtime implementation.  ...  Latecki and Mokhtarian for generously making contour data files available. Versions of Figures 1 and 3-5 have appeared in [30] [31] [32] . This research was supported in part by NSF 9802090.  ... 
doi:10.1142/s0218001406005174 fatcat:g3573wa36nfhtch2wnfthrjt6e

Improving object recognition accuracy and speed through nonuniform sampling

Boaz J. Super, David P. Casasent, Ernest L. Hall, Juha Roning
2003 Intelligent Robots and Computer Vision XXI: Algorithms, Techniques, and Active Vision  
Average time for retrieval and recognition is approximately 0.6 seconds in Matlab on a PC.  ...  In past work, we developed a fast and accurate method for retrieval and recognition of object silhouettes and other closed planar contours.  ...  Mokhtarian, Sclaroff, and Latecki for generously making contour data files available. The rabbit data 19 were derived from images in ref. [1] . Figures 1-4 are used by permission of CVRL.  ... 
doi:10.1117/12.519095 fatcat:tquvorvvzbbf7anhy7kikhb54q

Multi-View Combined Visual Similarity Retrieval for Face Database [chapter]

Yongsheng Gao, Maylor K. H. Leung
2002 Visual and Multimedia Information Management  
The search and retrieval employ a novel facial corner coding and matching approach, which delivers fast searching speed for database retrieval.  ...  This work proposed a new multi-view combined visual similarity retrieval approach for face database.  ...  The search and retrieval employ a novel facial comer coding and matching approach, which delivers fast searching speed for database retrieval.  ... 
doi:10.1007/978-0-387-35592-4_26 fatcat:bhzkwpso5rgwhf676svdcekxza

Hierarchical Local Maps for Robust Approximate Nearest Neighbor Computation

Pratyush Bhatt, Anoop M. Namboodiri
2009 2009 Seventh International Conference on Advances in Pattern Recognition  
In this paper, we propose a novel method for fast nearest neighbors retrieval in non-Euclidean and non-metric spaces.  ...  A method to find the approximate nearest neighbor of a query is proposed, that drastically reduces the total number of explicit distance measures that need to be computed.  ...  Hence the matching score cannot be used for computing approximate nearest neighbor.  ... 
doi:10.1109/icapr.2009.99 dblp:conf/icapr/BhattN09 fatcat:rwrwcvj7yzcldfb7khxbo6f3ea

Large-scale speaker identification

Ludwig Schmidt, Matthew Sharifi, Ignacio Lopez Moreno
2014 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
In this paper, we propose a system based on i-vectors, a current approach for speaker identification, and locality sensitive hashing, an algorithm for fast nearest neighbor search in high dimensions.  ...  the cosine distance in our retrieval procedure.  ...  CONCLUSION We have proposed a fast retrieval method for speaker identification in large data sets.  ... 
doi:10.1109/icassp.2014.6853878 dblp:conf/icassp/SchmidtSM14 fatcat:ziwnfmuqvzgrlefgv7aioymuba

Quantification of the optimal video-coding complexity for cost-efficient storage

E.G.T. Jaspers, J. Groenenboom
2005 2005 Digest of Technical Papers. International Conference on Consumer Electronics, 2005. ICCE.  
This paper describes the novelties of a video content analysis in a surveillance system, demonstrating the benefits for fast retrieval in huge video databases.  ...  Although many different types of technologies for information systems have evolved over the last decades (such as databases, video systems, the Internet and mobile telecommunication), the integration of  ...  Therefore, for each two MBR query result sets, a rank-join algorithm is executed that joins the trajectory points from the two sets.  ... 
doi:10.1109/icce.2005.1429748 fatcat:6pdocawlgnd5hgio35j3be7n3a

A non-linear dimensionality-reduction technique for fast similarity search in large databases

Khanh Vu, Kien A. Hua, Hao Cheng, Sheau-Dong Lang
2006 Proceedings of the 2006 ACM SIGMOD international conference on Management of data - SIGMOD '06  
In this reduction approach the approximation is unbounded, so that the approximation volume extends across the dataspace. This causes over-estimation of retrieval sets and impairs performance.  ...  To enable efficient similarity search in large databases, many indexing techniques use a linear transformation scheme to reduce dimensions and allow fast approximation.  ...  In response to a query, each technique returns a set of points that includes all qualifying points and some false matches.  ... 
doi:10.1145/1142473.1142532 dblp:conf/sigmod/VuHCL06 fatcat:pac6evebtfhm7nbsk7bhugmcwu

Fast Nearest-Neighbor Search Algorithms Based on High-Multidimensional Data

Rosslin John Robles
2013 Asia-pacific Journal of Multimedia services convergent with Art Humanities and Sociology  
Similarity search in multimedia databases requires an efficient support of nearest-neighbor search on a large set of high-dimensional points as a basic operation for query processing.  ...  uniform 23 distribution, the MBR approximations are identical with the NN-cells, which means that the approximations do not have any over-lap and therefore a point query on the index accesses only one  ...  for efficient nearest neighbor search in a set of high-dimensional points.  ... 
doi:10.14257/ajmscahs.2013.06.01 fatcat:t2gkyzewnbcsxmeatyy7orf3ym
« Previous Showing results 1 — 15 out of 119,588 results