Normalized Cuts and Image Segmentation

Normalized Cuts and Image Segmentation

ID:39508375

大小:2.75 MB

页数:18页

时间:2019-07-04

Normalized Cuts and Image Segmentation_第1页
Normalized Cuts and Image Segmentation_第2页
Normalized Cuts and Image Segmentation_第3页
Normalized Cuts and Image Segmentation_第4页
Normalized Cuts and Image Segmentation_第5页
资源描述:

《Normalized Cuts and Image Segmentation》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、888IEEETRANSACTIONSONPATTERNANALYSISANDMACHINEINTELLIGENCE,VOL.22,NO.8,AUGUST2000NormalizedCutsandImageSegmentationJianboShiandJitendraMalik,Member,IEEEAbstractÐWeproposeanovelapproachforsolvingtheperceptualgroupingprobleminvision.Ratherthanfocusingonlocalfea

2、turesandtheirconsistenciesintheimagedata,ourapproachaimsatextractingtheglobalimpressionofanimage.Wetreatimagesegmentationasagraphpartitioningproblemandproposeanovelglobalcriterion,thenormalizedcut,forsegmentingthegraph.Thenormalizedcutcriterionmeasuresboththe

3、totaldissimilaritybetweenthedifferentgroupsaswellasthetotalsimilaritywithinthegroups.Weshowthatanefficientcomputationaltechniquebasedonageneralizedeigenvalueproblemcanbeusedtooptimizethiscriterion.Wehaveappliedthisapproachtosegmentingstaticimages,aswellasmoti

4、onsequences,andfoundtheresultstobeveryencouraging.IndexTermsÐGrouping,imagesegmentation,graphpartitioning.æ1INTRODUCTIONNEARLY75yearsago,Wertheimer[24]pointedoutthetobedonefromthebigpicturedownward,ratherlikeaimportanceofperceptualgroupingandorganizationpaint

5、erfirstmarkingoutthemajorareasandthenfillingininvisionandlistedseveralkeyfactors,suchassimilarity,thedetails.proximity,andgoodcontinuation,whichleadtovisualPriorliteratureontherelatedproblemsofclustering,grouping.However,eventothisday,manyofthegroupingandimag

6、esegmentationishuge.Theclusteringcomputationalissuesofperceptualgroupinghavere-community[12]hasofferedusagglomerativeanddivisivemainedunresolved.Inthispaper,wepresentageneralalgorithms;inimagesegmentation,wehaveregion-basedframeworkforthisproblem,focusingspec

7、ificallyonthemergeandsplitalgorithms.Thehierarchicaldivisivecaseofimagesegmentation.approachthatweadvocateproducesatree,thedendrogram.SincetherearemanypossiblepartitionsofthedomainIWhilemostoftheseideasgobacktothe1970s(andearlier),ofanimageintosubsets,howdowe

8、picktheªrightºone?the1980sbroughtintheuseofMarkovRandomFields[10]Therearetwoaspectstobeconsideredhere.Thefirstisthatandvariationalformulations[17],[2],[14].TheMRFandtheremaynotbeasingleco

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。