Filters








21,232 Hits in 6.4 sec

Approximate search with quantized sparse representations [article]

Himalaya Jain, Patrick Pérez, Rémi Gribonval, Joaquin Zepeda and Hervé Jégou
2016 arXiv   pre-print
This paper tackles the task of storing a large collection of vectors, such as visual descriptors, and of searching in it.  ...  To this end, we propose to approximate database vectors by constrained sparse coding, where possible atom weights are restricted to belong to a finite subset.  ...  This classic vector quantization approach was recently used for approximate search [1, 8, 24] .  ... 
arXiv:1608.03308v1 fatcat:5ovfqxfr7bfsng56x6vf4v4z3m

Adaptive Cluster-Distance Bounding for Nearest Neighbor Search in Image Databases

Sharadh Ramaswamy, Kenneth Rose
2007 2007 IEEE International Conference on Image Processing  
We consider approaches for exact similarity search in a high dimensional space of correlated features representing image datasets, based on principles of clustering and vector quantization.  ...  Experiments conducted on real data-sets confirm the efficiency of our approach with random disk IOs reduced by 100X, as compared with the popular Vector Approximation-File (VA-File) approach, when allowed  ...  Navneet Panda, Vinay Melkote and Ankur Saxena for their valuable suggestions. The authors are extremely grateful to Balakrishnan Selvam for his help in performing the experiments.  ... 
doi:10.1109/icip.2007.4379601 dblp:conf/icip/RamaswamyR07 fatcat:ne7teonnlzgjna6c5vh7t4msyi

Approximate Search with Quantized Sparse Representations [chapter]

Himalaya Jain, Patrick Pérez, Rémi Gribonval, Joaquin Zepeda, Hervé Jégou
2016 Lecture Notes in Computer Science  
This paper tackles the task of storing a large collection of vectors, such as visual descriptors, and of searching in it.  ...  To this end, we propose to approximate database vectors by constrained sparse coding, where possible atom weights are restricted to belong to a finite subset.  ...  This classic vector quantization approach was recently used for approximate search [4, 5, 7] .  ... 
doi:10.1007/978-3-319-46478-7_42 fatcat:qogmfxb65ndcpmuwqtk2hzzjnu

4.-8. März 2019

Michael Günther, Maik Thiele, Wolfgang Lehner
2019 Datenbanksysteme für Business, Technologie und Web  
K nearest neighbor search (kNN-Search) is a universal data processing technique and a fundamental operation for word embeddings trained by word2vec or related approaches.  ...  In this paper, we develop an index structure for approximated kNN-Joins working well on high-dimensional data and provide an integration into PostgreSQL.  ...  and by the Intel® AI Research.  ... 
doi:10.18420/btw2019-15 dblp:conf/btw/GuntherTL19 fatcat:kjdfuowc5bagtgggrkyo2g34i4

Adaptable Similarity Search Using Vector Quantization [chapter]

Christian Böhm, Hans-Peter Kriegel, Thomas Seidl
2001 Lecture Notes in Computer Science  
Our solution provides a variety of new approximation techniques for quantized vectors which are combined by an extended multistep query processing architecture.  ...  Adaptable similarity queries based on quadratic form distance functions are widely popular in data mining applications, particularly for domains such as multimedia, CAD, molecular biology or medical image  ...  . for all : Vector Quantization.  ... 
doi:10.1007/3-540-44801-2_31 fatcat:jcgrx26ymbgv7luvxtc4vzxwxq

Affinity Preserving Quantization for Hashing: A Vector Quantization Approach to Learning Compact Binary Codes

Zhe Wang, Ling-Yu Duan, Tiejun Huang, Gao Wen
2016 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Hashing techniques are powerful for approximate nearest neighbour (ANN) search.Existing quantization methods in hashing are all focused on scalar quantization (SQ) which is inferior in utilizing the inherent  ...  methods, and has significantly improved the performance of various hashing techniques.  ...  Introduction Approximate nearest neighbour (ANN) search is widely used in machine learning, computer vision and information retrieval.  ... 
doi:10.1609/aaai.v30i1.10098 fatcat:g3y7qbidofgdfa7hare7wrblqi

Lucene for Approximate Nearest-Neighbors Search on Arbitrary Dense Vectors [article]

Tommaso Teofili, Jimmy Lin
2019 arXiv   pre-print
We demonstrate three approaches for adapting the open-source Lucene search library to perform approximate nearest-neighbor search on arbitrary dense vectors, using similarity search on word embeddings  ...  vectors, and k-d trees coupled with dimensionality reduction.  ...  Our demonstration explores techniques for performing approximate nearest-neighbor search on dense vectors directly in Lucene.  ... 
arXiv:1910.10208v2 fatcat:wuuh5do64nakhoumfzsf5xkebe

VQ-index

Ertem Tuncel, Hakan Ferhatosmanoglu, Kenneth Rose
2002 Proceedings of the tenth ACM international conference on Multimedia - MULTIMEDIA '02  
Our experiments demonstrate speedups of 20 to 35 over a VAfile technique that has been adapted for approximate nearest neighbor searching.  ...  In this paper, we introduce a novel indexing technique based on efficient compression of the feature space for approximate similarity searching in large multimedia databases.  ...  Scalar Quantization: Weber and Bohm [45] have proposed an approximate NN searching technique based on VAfiles.  ... 
doi:10.1145/641007.641117 dblp:conf/mm/TuncelFR02 fatcat:e6wpplvf7zgkxio3cptmqt4lmm

VQ-index

Ertem Tuncel, Hakan Ferhatosmanoglu, Kenneth Rose
2002 Proceedings of the tenth ACM international conference on Multimedia - MULTIMEDIA '02  
Our experiments demonstrate speedups of 20 to 35 over a VAfile technique that has been adapted for approximate nearest neighbor searching.  ...  In this paper, we introduce a novel indexing technique based on efficient compression of the feature space for approximate similarity searching in large multimedia databases.  ...  Scalar Quantization: Weber and Bohm [45] have proposed an approximate NN searching technique based on VAfiles.  ... 
doi:10.1145/641113.641117 fatcat:fukti4wdofcrtffqlffnil6m44

A New Partial Codeword Updating Scheme Based on Rate-Distortion Optimization for Adaptive Vector Quantization

Kai Guo, Lai-Man Po
2007 Multimedia and Expo, 2007 IEEE International Conference on  
Index Terms-Adaptive vector quantization, image and video coding, rate-distortion optimization  ...  The PCU-AVQ can be combined with transform coding and entropy coding for higher compression ratio, and it can be widely implemented in specific AVQ algorithms for image, video and speech coding.  ...  search technique.  ... 
doi:10.1109/icme.2007.4284759 dblp:conf/icmcs/GuoP07 fatcat:ggkkksgqljb2rff55f3yrjqmru

SAR IMAGE COMPRESSION USING ADAPTIVE DIFFERENTIAL EVOLUTION AND PATTERN SEARCH BASED K-MEANS VECTOR QUANTIZATION

Karri Chiranjeevi, Umaranjan Jena
2018 Image Analysis and Stereology  
A novel Vector Quantization (VQ) technique for encoding the Bi-orthogonal wavelet decomposed image using hybrid Adaptive Differential Evolution (ADE) and a Pattern Search optimization algorithm (hADEPS  ...  The superiority of proposed hybrid Adaptive Differential Evolution and Pattern Search (hADE-PS) optimized vector quantization over DE is demonstrated.  ...  VECTOR QUANTIZATION The vector quantization is a one of the block coding technique for image compression.  ... 
doi:10.5566/ias.1611 fatcat:smuorscjoraqje6ajmnu3sa7n4

Spreading vectors for similarity search [article]

Alexandre Sablayrolles, Matthijs Douze, Cordelia Schmid, Hervé Jégou
2019 arXiv   pre-print
State-of-the-art techniques learn parameters of quantizers on training data for optimal performance, thus adapting quantizers to the data.  ...  In this work, we propose to reverse this paradigm and adapt the data to the quantizer: we train a neural net which last layer forms a fixed parameter-free quantizer, such as pre-defined points of a hyper-sphere  ...  SIMILARITY SEARCH WITH LATTICE VECTOR QUANTIZERS We evaluate the lattice-based indexing proposed in Section 4, and compare it to more conventional methods based on quantization, namely PQ (Jégou et al  ... 
arXiv:1806.03198v3 fatcat:glvg7h5jqbfxdasbn5r4pxokqi

Anisotropic Additive Quantization for Fast Inner Product Search

Jin Zhang, Qi Liu, Defu Lian, Zheng Liu, Le Wu, Enhong Chen
2022 AAAI Conference on Artificial Intelligence  
To this end, we propose a quantization method called Anisotropic Additive Quantization to combine the scoreaware anisotropic loss and additive quantization.  ...  The experimental results show that it outperforms the stateof-the-art baselines with respect to approximate search accuracy while guaranteeing a similar retrieval efficiency.  ...  Acknowledgments The work was supported by grants from the National Natural Science Foundation of China (No. 62022077 and 61976198) and the Fundamental Research Funds for the Central Universities (No.  ... 
dblp:conf/aaai/ZhangLLLWC22 fatcat:bfkwrqnobjbmnk4a3rk2vjbypu

Product sub-vector quantization for feature indexing

The-Anh Pham, Dinh-Nghiep Le, Thi-Lan-Phuong Nguyen
2019 Journal of Computer Science and Cybernetics  
In addition, the quantized data can be jointly used with a clustering tree to perform approximate nearest search very efficiently.  ...  To this aim, we propose to perform product sub-vector quantization (PSVQ) to create finer representation of underlying data while still maintaining reasonable memory allocation.  ...  Recently, product quantization (PQ) [9] has been actively studied for its applications in fast approximate nearest neighbor search (ANN) and feature indexing.  ... 
doi:10.15625/1813-9663/35/1/13442 fatcat:vhqhs7oc55gg5az4prlla7r5fi

Image coding using vector quantization: a review

N.M. Nasrabadi, R.A. King
1988 IEEE Transactions on Communications  
This paper presents a review of vector quantization techniques used for encoding digital images.  ...  The emphasis here is on the usefulness of the vector quantization when it is combined with conventional image coding techniques or when it is used in different domains.  ...  Feng for drawing the block diagrams.  ... 
doi:10.1109/26.3776 fatcat:47zydmcogncnpfht7bzcm5z7x4
« Previous Showing results 1 — 15 out of 21,232 results