Filters








275 Hits in 4.5 sec

Boosting k-Nearest Neighbor Queries Estimating Suitable Query Radii

Marcos R. Vieira, Caetano Traina Jr., Agma J. M. Traina, Adriano Arantes, Christos Faloutsos
2007 International Conference on Scientific and Statistical Database Management  
This paper proposes novel and effective techniques to estimate a radius to answer k-nearest neighbor queries.  ...  The second technique targets datasets where the first technique cannot be employed, generating estimations that depend on where the query center is located.  ...  the k-Nearest Neighbor Query (kNNQ).  ... 
doi:10.1109/ssdbm.2007.5 dblp:conf/ssdbm/VieiraTTAF07 fatcat:dodruxqyurhhnkwm5r6cc2cyzq

Boosting k-NN for Categorization of Natural Scenes

Richard Nock, Paolo Piro, Frank Nielsen, Wafa Bel Haj Ali, Michel Barlaud
2012 International Journal of Computer Vision  
In this paper, we propose a novel boosting approach for generalizing the k-NN rule, by providing a new k-NN boosting algorithm, called UNN (Universal Nearest Neighbors), for the induction of leveraged  ...  The k-nearest neighbors (k-NN) classification rule has proven extremely successful in countless many computer vision applications.  ...  data in the feature space (kernel nearest neighbors); -distance-weighted and difference-weighted nearest neighbors; -boosting nearest neighbors.  ... 
doi:10.1007/s11263-012-0539-2 fatcat:bgp46fp3lfem7ovc7r7kddiacq

Learning Binary Hash Codes for Large-Scale Image Search [chapter]

Kristen Grauman, Rob Fergus
2013 Studies in Computational Intelligence  
Having done so, one can then search the data efficiently using hash tables, or by exploring the Hamming ball volume around a novel query.  ...  In particular, we review supervised methods that integrate metric learning, boosting, and neural networks into the hash key construction, and unsupervised methods based on spectral analysis or kernelized  ...  As such, the techniques are suitable for enhancing nearest neighbor categorization as well as similarity search for content-based retrieval.  ... 
doi:10.1007/978-3-642-28661-2_3 fatcat:7iird5mtw5h5rhp6xw3wugyuli

Approximate Nearest Neighbor Search on High Dimensional Data --- Experiments, Analyses, and Improvement (v1.0) [article]

Wen Li, Ying Zhang, Yifang Sun, Wei Wang, Wenjie Zhang, Xuemin Lin
2016 arXiv   pre-print
In this paper, we conduct a comprehensive experimental evaluation of many state-of-the-art methods for approximate nearest neighbor search.  ...  Approximate Nearest neighbor search (ANNS) is fundamental and essential operation in applications from many domains, such as databases, machine learning, multimedia, and computer vision.  ...  We also can define Relative Contrast for k-nearest neighbor setting as C k r = Dmean D knn , where D knn is the expected distance to the k-th nearest neighbor.  ... 
arXiv:1610.02455v1 fatcat:skn6iftztnhr7d524fqvyqix3m

Meridian

Bernard Wong, Aleksandrs Slivkins, Emin Gün Sirer
2005 Computer communication review  
The framework consists of an overlay network structured around multi-resolution rings, query routing with direct measurements, and gossip protocols for dissemination.  ...  In this space, we consider the N nearest-neighbor queries propagating through the Meridian network.  ...  Consider a node q and let u be its nearest neighbor. Say node v is a γ-approximate nearest neighbor of q if dvq/duq ≤ γ.  ... 
doi:10.1145/1090191.1080103 fatcat:am4s7pxgy5gnvacd74fkkm7y4q

Meridian

Bernard Wong, Aleksandrs Slivkins, Emin Gün Sirer
2005 Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications - SIGCOMM '05  
The framework consists of an overlay network structured around multi-resolution rings, query routing with direct measurements, and gossip protocols for dissemination.  ...  In this space, we consider the N nearest-neighbor queries propagating through the Meridian network.  ...  Consider a node q and let u be its nearest neighbor. Say node v is a γ-approximate nearest neighbor of q if dvq/duq ≤ γ.  ... 
doi:10.1145/1080091.1080103 dblp:conf/sigcomm/WongSS05 fatcat:hnmllebvdbchfoocw632m44ana

Diverse Yet Efficient Retrieval using Hash Functions [article]

Vidyadhar Rao, Prateek Jain, C.V Jawahar
2015 arXiv   pre-print
ball of a certain radii around the query point and also ensuring the diversity among the points.  ...  ball of a certain radii around the query point and also ensuring the diversity among the points.  ... 
arXiv:1509.06553v2 fatcat:gbfrtzxwvfdo3bbc3uxw7nnzma

Local shape feature fusion for improved matching, pose estimation and 3D object recognition

Anders G. Buch, Henrik G. Petersen, Norbert Krüger
2016 SpringerPlus  
In the 3D domain, local descriptors are an equally valuable mechanism for various estimation tasks, including object instance recognition and pose estimation.  ...  This work concerns the problem of selecting an optimal local feature for certain estimation tasks.  ...  For descriptor matching, it is therefore natural to reverse the direction, i.e. to find the nearest neighbor within the query model set for each of the target features.  ... 
doi:10.1186/s40064-016-1906-1 pmid:27066334 pmcid:PMC4783326 fatcat:fihapb36s5holaqs6ek2ehueza

Diverse Yet Efficient Retrieval using Locality Sensitive Hashing

Vidyadhar Rao, Prateek Jain, C.V. Jawahar
2016 Proceedings of the 2016 ACM on International Conference on Multimedia Retrieval - ICMR '16  
In our work, instead of finding exact nearest neighbors to a query, we retrieve approximate nearest neighbors that are diverse.  ...  ball of a certain radii around the query point and also ensuring the diversity among the points. 1.  ... 
doi:10.1145/2911996.2911998 dblp:conf/mir/RaoJJ16 fatcat:eb2lnpa2ovhtzdole5vp2o5wue

Beyond Cross-Validation—Accuracy Estimation for Incremental and Active Learning Models

Christian Limberg, Heiko Wersing, Helge Ritter
2020 Machine Learning and Knowledge Extraction  
We propose a novel semi-supervised accuracy estimation approach that clearly outperforms these two methods.  ...  We introduce the Configram Estimation (CGEM) approach to predict the accuracy of any classifier that delivers confidences.  ...  k-Nearest Neighbors (KNN) kNN is an instance-based classifier which simply memorizes samples L.  ... 
doi:10.3390/make2030018 fatcat:4hepw5mbtzc47nfwh4rzais7ge

A Survey on Trajectory Data Management, Analytics, and Learning [article]

Sheng Wang, Zhifeng Bao, J. Shane Culpepper, Gao Cong
2020 arXiv   pre-print
In this survey, we comprehensively review recent research trends in trajectory data management, ranging from trajectory pre-processing, storage, common trajectory analytic tools, such as querying spatial-only  ...  k Nearest Neighbors Query.  ...  An RkNN query aims to identify all (spatial) objects that have a query location as a k nearest neighbor.  ... 
arXiv:2003.11547v2 fatcat:5gf5h5skqjbrhf67cflygggnky

Breaking the polar-nonpolar division in solvation free energy prediction

Bao Wang, Chengzhang Wang, Kedi Wu, Guo-Wei Wei
2017 Journal of Computational Chemistry  
Moreover, for each target molecule, we adopt a machine learning algorithm for its nearest neighbor search, based on the selected microscopic feature vectors.  ...  Finally, from the extended feature vectors of obtained nearest neighbors, we construct a functional of solvation free energy, which is employed to predict the solvation free energy of the target molecule  ...  For nearest neighbor searches in each query, we emphasize that the nearest neighbor is measured based on the nearness of the solvation free energies, instead of the similarity measure used before.  ... 
doi:10.1002/jcc.25107 pmid:29127720 fatcat:4pga6lackbhjliixxvnetncmcy

Near neighbor searching with K nearest references

E. Chávez, M. Graff, G. Navarro, E.S. Téllez
2015 Information Systems  
A set of references is chosen from the database, and the signature of each object consists of the K references nearest to the object.  ...  In this paper we show that many of those recent indexes can be understood as variants of a simple general model based on K-nearest reference signatures.  ...  The most popular query in applications is that of retrieving the k nearest neighbors of a query, that is, the k database objects closest to it.  ... 
doi:10.1016/j.is.2015.02.001 fatcat:vjswiuixwvernlffgyfyatij5i

Performance of histogram descriptors for the classification of 3D laser range data in urban environments

Jens Behley, Volker Steinhage, Armin B. Cremers
2012 2012 IEEE International Conference on Robotics and Automation  
These descriptors are 1D, 2D, and 3D histograms capturing the distribution of normals or points around a query point.  ...  The selection of suitable features and their parameters for the classification of three-dimensional laser range data is a crucial issue for high-quality results.  ...  As all descriptors use a radius neighborhood N δ p , i.e., all points in a radius δ around the query point p instead of using the k-nearest neighbors N k p , we get a sampling invariant representation  ... 
doi:10.1109/icra.2012.6225003 dblp:conf/icra/BehleySC12 fatcat:p57jrpzkxveh5o25wfkutw6qbi

Partial Matching of 3D Shapes with Priority-Driven Search [article]

T. Funkhouser, P. Shilane
2006 Symposium on geometry processing : [proceedings]. Symposium on Geometry Processing  
of k features on the query match features on the target object.  ...  Given a query object and a database of target objects, all represented by sets of local 3D shape features, the algorithm produces a ranked list of the c best target objects sorted by how well any subset  ...  These descriptors can be searched efficiently, and thus they are suitable for queries into large databases of 3D shapes.  ... 
doi:10.2312/sgp/sgp06/131-142 fatcat:gkowtjcbaffb3jbifrih5bsdoy
« Previous Showing results 1 — 15 out of 275 results