Closest point search in high dimensions
The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 865 |
|---|---|
| container_issue | |
| container_start_page | 859 |
| container_title | |
| container_volume | |
| creator | Nene, S.A. Nayar, S.K. |
| description | The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user specified distance /spl epsiv/. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance /spl epsiv/. Our algorithm uses a projection search technique along with a novel data structure to dramatically improve performance in high dimensions. A complexity analysis is presented which can help determine /spl epsiv/ in structured problems. Benchmarks clearly show the superiority of the proposed algorithm for high dimensional search problems frequently encountered in machine vision, such as real-time object recognition. |
| doi_str_mv | 10.1109/CVPR.1996.517172 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>proquest_6IE</sourceid><recordid>TN_cdi_ieee_primary_517172</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>517172</ieee_id><sourcerecordid>26072160</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-e1cc09fc9e83f797d8b379e6d598632f79808b8bb8e80eba4ca0c74ae31a48713</originalsourceid><addsrcrecordid>eNpNkL1LxEAUxBc_wONML1aptEp8L5vs7isl-AUHiqht2GxevJVcErN3hf-9gVhYDcz8GIYR4gIhRQS6KT9eXlMkUmmBGnV2JFYISiaKkI5FRNqAQaN0VhCc_MvORBTCFwAgKYI8X4nrshsCh308Dr7fx4Ht5Lax7-Ot_9zGjd9xH_zQh3Nx2toucPSna_F-f_dWPiab54en8naT-AzkPmF0Dqh1xEa2mnRjaqmJVVOQUTKbrXlZberasAGube4sOJ1blmhzo1GuxdXSO07D92EeVu18cNx1tufhEKpMgc5QwQxeLqBn5mqc_M5OP9Vyh_wF9_pP9w</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype><pqid>26072160</pqid></control><display><type>conference_proceeding</type><title>Closest point search in high dimensions</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Nene, S.A. ; Nayar, S.K.</creator><creatorcontrib>Nene, S.A. ; Nayar, S.K.</creatorcontrib><description>The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user specified distance /spl epsiv/. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance /spl epsiv/. Our algorithm uses a projection search technique along with a novel data structure to dramatically improve performance in high dimensions. A complexity analysis is presented which can help determine /spl epsiv/ in structured problems. Benchmarks clearly show the superiority of the proposed algorithm for high dimensional search problems frequently encountered in machine vision, such as real-time object recognition.</description><identifier>ISSN: 1063-6919</identifier><identifier>ISBN: 9780818672590</identifier><identifier>ISBN: 0818672595</identifier><identifier>EISSN: 1063-6919</identifier><identifier>DOI: 10.1109/CVPR.1996.517172</identifier><language>eng</language><publisher>IEEE</publisher><subject>Computer science ; Computer vision ; Euclidean distance ; Face recognition ; Intelligent systems ; Machine vision ; Nearest neighbor searches ; Object recognition ; Partitioning algorithms ; Search problems</subject><ispartof>PROC IEEE COMPUT SOC CONF COMPUT VISION PATTERN RECOGNIT, 1996, p.859-865</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/517172$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2051,4035,4036,27904,54899</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/517172$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Nene, S.A.</creatorcontrib><creatorcontrib>Nayar, S.K.</creatorcontrib><title>Closest point search in high dimensions</title><title>PROC IEEE COMPUT SOC CONF COMPUT VISION PATTERN RECOGNIT</title><addtitle>CVPR</addtitle><description>The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user specified distance /spl epsiv/. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance /spl epsiv/. Our algorithm uses a projection search technique along with a novel data structure to dramatically improve performance in high dimensions. A complexity analysis is presented which can help determine /spl epsiv/ in structured problems. Benchmarks clearly show the superiority of the proposed algorithm for high dimensional search problems frequently encountered in machine vision, such as real-time object recognition.</description><subject>Computer science</subject><subject>Computer vision</subject><subject>Euclidean distance</subject><subject>Face recognition</subject><subject>Intelligent systems</subject><subject>Machine vision</subject><subject>Nearest neighbor searches</subject><subject>Object recognition</subject><subject>Partitioning algorithms</subject><subject>Search problems</subject><issn>1063-6919</issn><issn>1063-6919</issn><isbn>9780818672590</isbn><isbn>0818672595</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>1996</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNpNkL1LxEAUxBc_wONML1aptEp8L5vs7isl-AUHiqht2GxevJVcErN3hf-9gVhYDcz8GIYR4gIhRQS6KT9eXlMkUmmBGnV2JFYISiaKkI5FRNqAQaN0VhCc_MvORBTCFwAgKYI8X4nrshsCh308Dr7fx4Ht5Lax7-Ot_9zGjd9xH_zQh3Nx2toucPSna_F-f_dWPiab54en8naT-AzkPmF0Dqh1xEa2mnRjaqmJVVOQUTKbrXlZberasAGube4sOJ1blmhzo1GuxdXSO07D92EeVu18cNx1tufhEKpMgc5QwQxeLqBn5mqc_M5OP9Vyh_wF9_pP9w</recordid><startdate>1996</startdate><enddate>1996</enddate><creator>Nene, S.A.</creator><creator>Nayar, S.K.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>1996</creationdate><title>Closest point search in high dimensions</title><author>Nene, S.A. ; Nayar, S.K.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-e1cc09fc9e83f797d8b379e6d598632f79808b8bb8e80eba4ca0c74ae31a48713</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>1996</creationdate><topic>Computer science</topic><topic>Computer vision</topic><topic>Euclidean distance</topic><topic>Face recognition</topic><topic>Intelligent systems</topic><topic>Machine vision</topic><topic>Nearest neighbor searches</topic><topic>Object recognition</topic><topic>Partitioning algorithms</topic><topic>Search problems</topic><toplevel>online_resources</toplevel><creatorcontrib>Nene, S.A.</creatorcontrib><creatorcontrib>Nayar, S.K.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Nene, S.A.</au><au>Nayar, S.K.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Closest point search in high dimensions</atitle><btitle>PROC IEEE COMPUT SOC CONF COMPUT VISION PATTERN RECOGNIT</btitle><stitle>CVPR</stitle><date>1996</date><risdate>1996</risdate><spage>859</spage><epage>865</epage><pages>859-865</pages><issn>1063-6919</issn><eissn>1063-6919</eissn><isbn>9780818672590</isbn><isbn>0818672595</isbn><abstract>The problem of finding the closest point in high-dimensional spaces is common in computational vision. Unfortunately, the complexity of most existing search algorithms, such as k-d tree and R-tree, grows exponentially with dimension, making them impractical for dimensionality above 15. In nearly all applications, the closest point is of interest only if it lies within a user specified distance /spl epsiv/. We present a simple and practical algorithm to efficiently search for the nearest neighbor within Euclidean distance /spl epsiv/. Our algorithm uses a projection search technique along with a novel data structure to dramatically improve performance in high dimensions. A complexity analysis is presented which can help determine /spl epsiv/ in structured problems. Benchmarks clearly show the superiority of the proposed algorithm for high dimensional search problems frequently encountered in machine vision, such as real-time object recognition.</abstract><pub>IEEE</pub><doi>10.1109/CVPR.1996.517172</doi><tpages>7</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1063-6919 |
| ispartof | PROC IEEE COMPUT SOC CONF COMPUT VISION PATTERN RECOGNIT, 1996, p.859-865 |
| issn | 1063-6919 1063-6919 |
| language | eng |
| recordid | cdi_ieee_primary_517172 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Computer science Computer vision Euclidean distance Face recognition Intelligent systems Machine vision Nearest neighbor searches Object recognition Partitioning algorithms Search problems |
| title | Closest point search in high dimensions |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-23T03%3A25%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Closest%20point%20search%20in%20high%20dimensions&rft.btitle=PROC%20IEEE%20COMPUT%20SOC%20CONF%20COMPUT%20VISION%20PATTERN%20RECOGNIT&rft.au=Nene,%20S.A.&rft.date=1996&rft.spage=859&rft.epage=865&rft.pages=859-865&rft.issn=1063-6919&rft.eissn=1063-6919&rft.isbn=9780818672590&rft.isbn_list=0818672595&rft_id=info:doi/10.1109/CVPR.1996.517172&rft_dat=%3Cproquest_6IE%3E26072160%3C/proquest_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=26072160&rft_id=info:pmid/&rft_ieee_id=517172&rfr_iscdi=true |