CV大牛/研究组/博客/算法/代码/数据集|资源地图|超干货必收藏

牛人Homepages(随意排序,不分先后):

1.USCComputerVisionGroup:南加大,多目标跟踪/检测等;

2.ETHZComputerVisionLaboratory:苏黎世联邦理工学院,欧洲最好的几个CV/ML研究机构;

3.HelmutGrabner:OnlineBoostingandVision的作者,trackingbyonlinefeatureselection的早期经典,貌似现在不是很活跃了,跑去创业了;

4.:PSU,也是跟踪界的大牛;

5.YingWu:美国西北大学,华人学者中的翘楚;

6.JunsongYuan:NTU,上面Wu老师的学生;

8.TheAustralianCentreforVisualTechnologies:阿德莱德大学的CV组,最近也是exceedinglyactivefruitful;

9.ChunhuaShen:属上面的ACVT组,最近非常活跃;

10.XiLi:同属ACVT,之前是中科院的PHD,跟踪方面的论文很多,有理论深度;

11.HaibinLing:天普大学,L1-Tracker及后续扩展,源码分享;

12.Learning,Recognition,andSurveillance:奥地利TUGraz,在线学习,跟踪/检测等,active!源码分享;

13.StatisticalVisualComputingLaboratory:UCSD,光听名字就很学术吧,Saliency研究很有名;

14.DavidRoss:多伦多大学,IVT的作者,跟踪中Generative表观的经典中的经典,提供源码,IVT的代码结构被后来很多人引用,值得一读;

15.EPFL,ComputerVisionLaboratory:洛桑理工的学院,和上面的的ETHZCVlab同样是欧洲最好的CV研究大组;

16.JamieShotton:属微软剑桥研究中心,Decision/RegressionForests;

17.SinisaTodorovic:俄勒冈州立,行为分析等;

18.ShiJianbo:大名鼎鼎的GoodFeaturetoTrack作者,目前方向行为分析和多目标跟踪等;

19.ShaiAvidan:特拉维夫大学,大牛级,可算是Tracking-by-detection的开创者,EnsembleTracking,SVMTracking;

20.VisualInformationProcessingandLearning:中科院计算所,山世光老师的研究组,不需介绍了吧;

21.ShaogangGong:QueenMaryUniversityofLondon,各种PAMI,IJCV;

22.YangJian:南京理工大学,2DPCA,人脸识别;

23.CALVIN:weaklysupervisedlearning,objectness;

24.LearningVisionGroup:NUS,稀疏表示;

26.XiaogangWang:CUHK,activefruitful,行人检测,群体行为分析;

27.Zhou,Bolei:上面Wang老师硕士研究生,群体行为,看看人家的Publications已经轻松甩国内博士好几条街;

28.ComputationalVisionGroup:Leader--DevaRamanan;

29.ZhangLei:香港理工,稀疏表示,人脸识别,可以算大中华区比较活跃的研究组了,几乎每篇论文都有对应源码;

30.ZhangKaihua:上面Zhang老师学生,CompressiveTracking;

31.PramodSharma:离线训练检测器的在线自适应,貌似是个不错的topic;

32.LorisBazzani:personre-id,他的SDALF(code)描述子经常被用来做为比较对象,说明还是有参考价值的;

33.PedroFelzenszwalb:布朗大学,目标检测,新新N人一枚;

34.VijayakumarBhagavatula:IEEEFellow,correlationfilters;

35.LaurensvanderMaaten:MLer.

牛人主页(主页有很多论文代码)

SergeBelongieatUCSanDiego

AntonioTorralbaatMIT

AlexeiFfrosatCMU

CeLiuatMicrosoftResearchNewEngland

KristenGraumanatUTAustin

DeviParikhatTTI-Chicago(MarrPrizeatICCV2011)

JohnWrightatColumbiaUniv.

PiotrDollaratCalTech

BorisBabenkoatUCSanDiego

DavidRossatGoogle/Youtube

DavidDonohoatStanfordUniv.

大神们:

RobertoCipollaatCambridge

YiMaatMSRA

重要研究组:

ComputerVisionGroupatUCBerkeley

LEARatINRIA

ComputerVisionLabatStanford

ComputerVisionLabatEPFL

ComputerVisionLabatETHZurich

ComputerVisionLabatSeoulNationalUniv.

ComputerVisionLabatUCSanDiego

ComputerVisionLabatUCSantaCruz

ComputerVisionLabatColumbiaUniv.

UCLAVisionLab

MotionandShapeComputingGroupatGeorgeMasonUniv.

RobustImageUnderstandingLabatRutgersUniv.

ahref="http://per"PersonalRoboticsLabatCMU

VisualPerceptionLabatPurdueUniv.

潜力牛人:

JuergenGallatETHZurich

MattFlaggatGeorgiaTech.

MathieuSalzmannatTTI-Chicago

GergShakhnarovichatTTI-Chicago

TaegSangChoatMIT

JianchaoYangatUIUC

StefanRothatTUDarmstadt

YinanYuatCASIA(PASCALVOC2010DetectionChallengeWinner)

ZdenekKalalatFPFL

JulienPiletatFPFL

KenjiOkuma

个人、研究机构链接

(1)googleResearch;

(2)MIT博士,汤晓欧学生林达华;

(3)MIT博士后DouglasLanman;

(4)opencv中文网站;

(5)Stanford大学vision实验室;

(6)Stanford大学博士崔靖宇;

(7)UCLA教授朱松纯;

(8)中国人工智能网;

(9)中国视觉网;

(10)中科院自动化所;

(11)中科院自动化所李子青研究员;

(12)中科院计算所山世光研究员;

(13)人脸识别主页;

(14)加州大学伯克利分校CV小组;

(15)南加州大学CV实验室;

(16)卡内基梅隆大学CV主页;

(17)微软CV研究员RichardSzeliski;

(18)微软亚洲研究院计算机视觉研究组;

(19)微软剑桥研究院ML与CV研究组;

(20)研学论坛;

(21)美国Rutgers大学助理教授刘青山;

(22)计算机视觉最新资讯网;

(24)香港中文大学助理教授王晓刚;

(25)香港中文大学多媒体实验室(汤晓鸥);

(26);

(27)CVonline;

(28)computervisionsoftware;

(29)ComputerVisionResource;

(30)computervisionresearchgroups;

(31)computervisioncenter;

(32)浙江大学图像技术研究与应用(ITRA)团队:

(33)自动识别网:

(34)清华大学章毓晋教授:

(35)顶级民用机器人研究小组领导的WillowGarage:

(36)上海交通大学图像处理与模式识别研究所:

(37)上海交通大学计算机视觉实验室刘允才教授:

(38)德克萨斯州大学奥斯汀分校助理教授KristenGrauman:,检索

(39)清华大学电子工程系智能图文信息处理实验室(丁晓青教授):

(40)北京大学高文教授:

(41)清华大学艾海舟教授:

(42)中科院生物识别与安全技术研究中心:

(43)瑞士巴塞尔大学ThomasVetter教授:

(44)俄勒冈州立大学RobHess博士:

(45)深圳大学于仕祺副教授:

(46)西安交通大学人工智能与机器人研究所:

(47)卡内基梅隆大学研究员:

(48)MIT博士ChrisStauffer:

(49)美国密歇根州立大学生物识别研究组(教授):

(50)美国伊利诺伊州立大学:

(51)武汉大学数字摄影测量与计算机视觉研究中心:

(52)瑞士巴塞尔大学SamiRomdhani助理研究员:

(53)CMU大学研究员YangWang:

(54)英国曼彻斯特大学TimCootes教授:

(55)美国罗彻斯特大学教授JieboLuo:

(56)美国普渡大学机器人视觉实验室:

(57)美国宾利州立大学感知、运动与认识实验室:

(58)美国宾夕法尼亚大学GRASP实验室:

(59)美国内达华大学里诺校区CV实验室:

(60)美国密西根大学vision实验室:

(61)UniversityofMassachusetts(麻省大学),视觉实验室:

(62)华盛顿大学博士后IvaKemelmacher:

(63)以色列魏茨曼科技大学RonenBasri:

(64)瑞士ETH-Zurich大学CV实验室:

(65)微软CV研究员张正友:

(66)中科院自动化所医学影像研究室:

(67)中科院田捷研究员:

(68)微软Redmond研究院研究员SimonBaker:

(69)普林斯顿大学教授李凯:

(70)普林斯顿大学博士贾登:

(71)牛津大学教授AndrewZisserman:

(72)英国leeds大学研究员MarkEveringham:

(73)英国爱丁堡大学教授ChrisWilliam:

(74)微软剑桥研究院研究员JohnWinn:

(75)佐治亚理工学院教授:

(76)微软亚洲研究院研究员孙剑:

(77)微软亚洲研究院研究员马毅:

(78)英国哥伦比亚大学教授DavidLowe:

(79)英国爱丁堡大学教授BobFisher:

(80)加州大学圣地亚哥分校教授:

(81)威斯康星大学教授:

(82)多伦多大学教授:

(83)伦斯勒理工学院教授QiangJi:

(84)CMU研究员DanielHuber:

(85)多伦多大学教授::

(86)伦敦大学玛丽女王学院教授AndreaCavallaro:

(87)多伦多大学教授KyrosKutulakos:

(88)杜克大学教授CarloTomasi:

(89)CMU教授MartialHebert:

(90)MIT助理教授AntonioTorralba:

(91)马里兰大学研究员YaselYacoob:

(92)康奈尔大学教授RaminZabih:

(93)CMU博士田渊栋:

(94)CMU副教授SrinivasaNarasimhan:

(95)CMU大学ILIM实验室:

(96)哥伦比亚大学教授:

(97)三菱电子研究院研究员FatihPorikli:

(98)康奈尔大学教授DanielHuttenlocher:

(99)南京大学教授周志华:

(100)芝加哥丰田技术研究所助理教授DeviParikh:

(101)瑞士联邦理工学院博士后HelmutGrabner:

(102)香港中文大学教授贾佳亚:

(103)南京大学教授吴建鑫:

(104)GE研究院研究员李关:

(105)佐治亚理工学院教授MonsonHayes:

(106)图片检索国际竞赛PASCALVOC(微软剑桥研究院组织):

(107)机器视觉开源处理库汇总:

(108)布朗大学教授BenjaminKimia:

(109)数据堂-图像处理相关的样本数据:

(110)东软基于CV的汽车辅助驾驶系统:

(111)马里兰大学教授RemaChellappa:

(112)芝加哥丰田研究中心助理教授DeviParikh:

(113)宾夕法尼亚大学助理教授石建波:

(114)比利时鲁汶大学教授LucVanGool:

(115)行人检测主页:

(116)法国学习算法与系统实验室BasilioNoris博士:

(117)美国马里兰大学教授:

(118)计算机视觉论文分类导航:

(119)计算机视觉分类信息导航:

(120)西班牙马德里理工大学博士MarcosNieto:

(121)香港理工大学副教授张磊:

(122)以色列技术学院教授MichaelElad:

(123)韩国启明大学计算机视觉与模式识别实验室:

(124)英国诺丁汉大学MichelValstar博士:

(125)卡内基梅隆大学TakeoKanade教授:

(126)微软学术搜索:

(127)比利时天主教鲁汶大学RaduTimofte博士:,交通标志检测,定位,3D跟踪

(128)迪斯尼匹兹堡研究院研究员:IainMatthews:

;person_id=741AAM,三维重建

(129)康奈尔大学视觉与图像分析组:

(130)密西根州立大学生物识别研究组:、指纹识别、图像检索

(131)柏林科技大学计算机视觉与遥感实验室:、物体重建、基于图像的表面测量、医学图像处理

(134)北卡莱罗纳大学教堂山分校MarcPollefeys教授:、相机标定、运动检测与分析、3D重建

(135)澳大利亚国立大学RichardHartley教授:、稀疏子空间、跟踪、

(136)百度技术副总监于凯:,稀疏表示,图像分类

(137)西安电子科技大学高新波教授:、水印、稀疏表示、超分辨率

(138)加州大学伯克利分校教授:

(139)加州理工行人检测相关资料:

(140)微软Redmond研究院研究员PiotrDollar:、特征提取、

(141)视觉计算研究论坛:

(142)美国坦桑尼亚州立大学稀疏学习软件包:

(143)美国加州大学圣地亚哥分校JacobWhitehill博士:

(144)美国布朗大学教授:

(145)美国加州大学圣地亚哥分校DavidKriegman教授:

(146)南加州大学PaulDebevec教授:

(147)伊利诺伊大学教授:

(148)英国牛津大学IanReid教授:

(149)CMU大学AlyoshaEfros教授:

(151)MIT教授WilliamFreeman:

(152)CMU博士HenrySchneiderman:;

(153)微软研究员PaulViola:

(154)微软研究员AntonioCriminisi:,三维重建,目标检测与跟踪;

(155)魏茨曼科学研究所教授MichalIrani:

(156)瑞士洛桑理工学院PascalFua教授:,增强现实

(157)佐治亚理工学院IrfanEssa教授:

(158)中科院助理教授樊彬:;

(159)斯坦福大学SebastianThrun教授:;

(160)多伦多大学教授:

(161)凤巢系统架构师张栋博士:

(162)2012年龙星计划机器学习课程:

(163)中科院自动化所肖柏华教授:、人脸识别、质量评判

(165)纽约大学YannLeCun教授

(166)二维条码识别开源库zxing:

(167)布朗大学PedroFelzenszwalb教授:,DeformablePartModel

(168)伊利诺伊香槟大学SvetlanaLazebnik教授:,聚类,图像检索

(169)荷兰乌德勒支大学图像与多媒体研究中心、多媒体检索与匹配

(171)中科院自动化所孙哲南助理教书:、掌纹识别、人脸识别

(172)南京信息工程大学刘青山教授:、医学图像分析

(173)清华大学助理教授冯建江:

(174)北航助理教授黄迪:

(175)中山大学助理教授郑伟诗:、特征匹配、聚类、检索;

(176)google瑞士苏黎世的工程师ThomasDeselaers:

(177)百度深度学习研究中心博士后余轶南:,图像检索

(178)威兹曼科技大学超分辨率:

(179)德克萨斯大学奥斯汀分校AlBovik教授:、特征提取

(180)以色列希伯来大学YairWeiss教授:、超分辨率

(181)以色列希伯来大学DanielZoran博士:、去噪

(182)美国加州大学PeymanMilanfar教授:

(183)中科院计算所副研究员常虹:、半监督学习、超分辨率

(184)以色列威茨曼大学AnatLevin教授:、去模糊

(185)以色列威茨曼大学DanielGlasner博士后:、分割、姿态估计

(186)密西根大学助理教授HonglakLee:、特征提取,去噪、稀疏表示;

(187)MIT周博磊博士:、运动检测

(188)美国田纳西大学LiHe博士:、超分辨率;

(189)Adobe研究院JianchaoYang研究员:,超分辨率、图片检索、去噪、去模糊

(190)DeepLearning主页:、软件,代码,demo,数据等;

(191)斯坦福大学AndrewNg教授:,深度学习

(192)Elefant:

(193)微软研究员CeLiu:、超分辨率、去模糊、分割

(194)WestVirginia大学助理教授XinLi:、降噪、去模糊

(195)、去噪、编码、压缩感知、超分辨率、聚类、分割等相关代码集合

(196)西班牙格拉纳达大学超分辨率重建项目组:

(197)清华大学程明明博士:、检索

(198)牛津布鲁克斯大学教授:、三维重建

(199)佐治亚理工学院教授:、行人检测、特征描述、

(200)大规模图像分类、检测竞赛ILSVRC(Stanford,Google举办):

(201)加州大学尔湾分校DevaRamanan助理教授:,行人检测,跟踪、稀疏表示

(202)人脸识别测试图片集:

(203)美国西北大学博士MingYang:、图像检索;

(204)美国加州大学伯克利分校博士后:(DPM)

(205)中文语言资源联盟:、字符识别的训练,测试库;

(206)西班牙巴塞罗那大学计算机视觉中心:、跟踪、3D、行人检测、汽车辅助驾驶

(207)德国戴姆勒研究所:、行人检测、

(208)苏黎世联邦理工学院AndreasEss博士后:、行为检测、跟踪

(209)Libqrencode:

(210)江西财经大学袁飞牛教授:、3D重建、医学图像处理

(211)耶路撒冷大学RaananFattal教师:、

(212)耶路撒冷大学DaniLischnski教授:、纹理合成、图像增强

代码汇总

一、特征提取FeatureExtraction:

SIFT[1][Demoprogram][SIFTLibrary][VLFeat]

PCA-SIFT[2][Project]

Affine-SIFT[3][Project]

SURF[4][OpenSURF][MatlabWrapper]

AffineCovariantFeatures[5][Oxfordproject]

MSER[6][Oxfordproject][VLFeat]

GeometricBlur[7][Code]

LocalSelf-SimilarityDescriptor[8][Oxfordimplementation]

GlobalandEfficientSelf-Similarity[9][Code]

HistogramofOrientedGraidents[10][INRIAObjectLocalizationToolkit][OLTtoolkitforWindows]

GIST[11][Project]

ShapeContext[12][Project]

ColorDescriptor[13][Project]

PyramidsofHistogramsofOrientedGradients[Code]

Space-TimeInterestPoints(STIP)[14][Project][Code]

BoundaryPreservingDenseLocalRegions[15][Project]

WeightedHistogram[Code]

Histogram-basedInterestPointsDetectors[Paper][Code]

AnOpenCV-C++implementationofLocalSelfSimilarityDescriptors[Project]

FastSparseRepresentationwithPrototypes[Project]

CornerDetection[Project]

AGASTCornerDetector:fasterthanFASTandevenFAST-ER[Project]

Real-timeFacialFeatureDetectionusingConditionalRegressionForests[Project]

GlobalandEfficientSelf-SimilarityforObjectClassificationandDetection[code]

WαSH:Weightedα-ShapesforLocalFeatureDetection[Project]

HOG[Project]

OnlineSelectionofDiscriminativeTrackingFeatures[Project]

二、图像分割ImageSegmentation:

NormalizedCut[1][Matlabcode]

GergMori’Superpixelcode[2][Matlabcode]

EfficientGraph-basedImageSegmentation[3][C++code][Matlabwrapper]

Mean-ShiftImageSegmentation[4][EDISONC++code][Matlabwrapper]

OWT-UCMHierarchicalSegmentation[5][Resources]

Turbepixels[6][Matlabcode32bit][Matlabcode64bit][Updatedcode]

Quick-Shift[7][VLFeat]

SLICSuperpixels[8][Project]

SegmentationbyMinimumCodeLength[9][Project]

BiasedNormalizedCut[10][Project]

SegmentationTree[11-12][Project]

EntropyRateSuperpixelSegmentation[13][Code]

FastApproximateEnergyMinimizationviaGraphCuts[Paper][Code]

EfficientPlanarGraphCutswithApplicationsinComputerVision[Paper][Code]

IsoperimetricGraphPartitioningforImageSegmentation[Paper][Code]

RandomWalksforImageSegmentation[Paper][Code]

BlossomV:Anewimplementationofaminimumcostperfectmatchingalgorithm[Code]

AnExperimentalComparisonofMin-Cut/Max-FlowAlgorithmsforEnergyMinimizationinComputerVision[Paper][Code]

GeodesicStarConvexityforInteractiveImageSegmentation[Project]

ContourDetectionandImageSegmentationResources[Project][Code]

BiasedNormalizedCuts[Project]

Max-flow/min-cut[Project]

Chan-VeseSegmentationusingLevelSet[Project]

AToolboxofLevelSetMethods[Project]

Re-initializationFreeLevelSetEvolutionviaReactionDiffusion[Project]

ImprovedC-Vactivecontourmodel[Paper][Code]

AVariationalMultiphaseLevelSetApproachtoSimultaneousSegmentationandBiasCorrection[Paper][Code]

LevelSetMethodResearchbyChunmingLi[Project]

ClassCutforUnsupervisedClassSegmentation[code]

SEEDS:SuperpixelsExtractedviaEnergy-DrivenSampling[Project][other]

三、目标检测ObjectDetection:

Asimpleobjectdetectorwithboosting[Project]

INRIAObjectDetectionandLocalizationToolkit[1][Project]

DiscriminativelyTrainedDeformablePartModels[2][Project]

CascadeObjectDetectionwithDeformablePartModels[3][Project]

Poselet[4][Project]

ImplicitShapeModel[5][Project]

ViolaandJones’sFaceDetection[6][Project]

BayesianModellingofDyanmicScenesforObjectDetection[Paper][Code]

Handdetectionusingmultipleproposals[Project]

ColorConstancy,IntrinsicImages,andShapeEstimation[Paper][Code]

Discriminativelytraineddeformablepartmodels[Project]

GradientResponseMapsforReal-TimeDetectionofTexture-LessObjects:LineMOD[Project]

ImageProcessingOnLine[Project]

RobustOpticalFlowEstimation[Project]

Where'sWaldo:MatchingPeopleinImagesofCrowds[Project]

ScalableMulti-classObjectDetection[Project]

Class-SpecificHoughForestsforObjectDetection[Project]

DeformedLatticeDetectionInReal-WorldImages[Project]

Discriminativelytraineddeformablepartmodels[Project]

四、显著性检测SaliencyDetection:

Itti,Koch,andNiebur’saliencydetection[1][Matlabcode]

Frequency-tunedsalientregiondetection[2][Project]

Saliencydetectionusingmaximumsymmetricsurround[3][Project]

AttentionviaInformationMaximization[4][Matlabcode]

Context-awaresaliencydetection[5][Matlabcode]

Graph-basedvisualsaliency[6][Matlabcode]

Saliencydetection:Aspectralresidualapproach.[7][Matlabcode]

Segmentingsalientobjectsfromimagesandvideos.[8][Matlabcode]

SaliencyUsingNaturalstatistics.[9][Matlabcode]

DiscriminantSaliencyforVisualRecognitionfromClutteredScenes.[10][Code]

LearningtoPredictWhereHumansLook[11][Project]

GlobalContrastbasedSalientRegionDetection[12][Project]

BayesianSaliencyviaLowandMidLevelCues[Project]

Top-DownVisualSaliencyviaJointCRFandDictionaryLearning[Paper][Code]

SaliencyDetection:ASpectralResidualApproach[Code]

五、图像分类、聚类ImageClassification,Clustering

PyramidMatch[1][Project]

SpatialPyramidMatching[2][Code]

Locality-constrainedLinearCoding[3][Project][Matlabcode]

SparseCoding[4][Project][Matlabcode]

TextureClassification[5][Project]

MultipleKernelsforImageClassification[6][Project]

FeatureCombination[7][Project]

SuperParsing[Code]

LargeScaleCorrelationClusteringOptimization[Matlabcode]

DetectingandSketchingtheCommon[Project]

Self-TuningSpectralClustering[Project][Code]

UserAssistedSeparationofReflectionsfromaSingleImageUsingaSparsityPrior[Paper][Code]

FiltersforTextureClassification[Project]

MultipleKernelLearningforImageClassification[Project]

SLICSuperpixels[Project]

六、抠图ImageMatting

AClosedFormSolutiontoNaturalImageMatting[Code]

SpectralMatting[Project]

Learning-basedMatting[Code]

七、目标跟踪ObjectTracking:

AForestofSensors-TrackingAdaptiveBackgroundMixtureModels[Project]

ObjectTrackingviaPartialLeastSquaresAnalysis[Paper][Code]

RobustObjectTrackingwithOnlineMultipleInstanceLearning[Paper][Code]

OnlineVisualTrackingwithHistogramsandArticulatingBlocks[Project]

IncrementalLearningforRobustVisualTracking[Project]

Real-timeCompressiveTracking[Project]

RobustObjectTrackingviaSparsity-basedCollaborativeModel[Project]

VisualTrackingviaAdaptiveStructuralLocalSparseAppearanceModel[Project]

OnlineDiscriminativeObjectTrackingwithLocalSparseRepresentation[Paper][Code]

SuperpixelTracking[Project]

LearningHierarchicalImageRepresentationwithSparsity,SaliencyandLocality[Paper][Code]

OnlineMultipleSupportInstanceTracking[Paper][Code]

VisualTrackingwithOnlineMultipleInstanceLearning[Project]

Objectdetectionandrecognition[Project]

CompressiveSensingResources[Project]

RobustReal-TimeVisualTrackingusingPixel-WisePosteriors[Project]

Tracking-Learning-Detection[Project][OpenTLD/C++Code]

theHandVu:vision-basedhandgestureinterface[Project]

LearningProbabilisticNon-LinearLatentVariableModelsforTrackingComplexActivities[Project]

八、Kinect:

Kinecttoolbox[Project]

OpenNI[Project]

zouxy09CSDNBlog[Resource]

FingerTracker手指跟踪[code]

九、3D相关:

3DReconstructionofaMovingObject[Paper][Code]

ShapeFromShadingUsingLinearApproximation[Code]

CombiningShapefromShadingandStereoDepthMaps[Project][Code]

ShapefromShading:ASurvey[Paper][Code]

ASpatio-TemporalDescriptorbasedon3DGradients(HOG3D)[Project][Code]

Multi-cameraSceneReconstructionviaGraphCuts[Paper][Code]

AFastMarchingFormulationofPerspectiveShapefromShadingunderFrontalIllumination[Paper][Code]

Reconstruction:3DShape,Illumination,Shading,Reflectance,Texture[Project]

MonocularTrackingof3DHumanMotionwithaCoordinatedMixtureofFactorAnalyzers[Code]

Learning3-DSceneStructurefromaSingleStillImage[Project]

十、机器学习算法:

MatlabclassforcomputingApproximateNearestNieghbor(ANN)[MatlabclassprovidinginterfacetoANNlibrary]

RandomSampling[code]

ProbabilisticLatentSemanticAnalysis(pLSA)[Code]

FASTANNandFASTCLUSTERforapproximatek-means(AKM)[Project]

FastIntersection/AdditiveKernelSVMs[Project]

SVM[Code]

Ensemblelearning[Project]

DeepLearning[Net]

DeepLearningMethodsforVision[Project]

NeuralNetworkforRecognitionofHandwrittenDigits[Project]

TrainingadeepautoencoderoraclassifieronMNISTdigits[Project]

THEMNISTDATABASEofhandwrittigits[Project]

Ersatz:deepneuralnetworksinthecloud[Project]

DeepLearning[Project]

sparseLM:SparseLevenberg-MarquardtnonlinearleastsquaresinC/C++[Project]

Weka3:DataMiningSoftwareinJava[Project]

Invitedtalk"ATutorialonDeepLearning"(余凯)[Video]

CNN-Convolutionalneuralnetworkclass[MatlabTool]

YannLeCun'sPublications[Wedsite]

LeNet-5,convolutionalneuralnetworks[Project]

TrainingadeepautoencoderoraclassifieronMNISTdigits[Project]

DeepLearning大牛'sHomePage[Website]

MultipleInstanceLogisticDiscriminant-basedMetricLearning(MildML)andLogisticDiscriminant-basedMetricLearning(LDML)[Code]

Sparsecodingsimulationsoftware[Project]

VisualRecognitionandMachineLearningSummerSchool[Software]

十一、目标、行为识别Object,ActionRecognition:

ActionRecognitionbyDenseTrajectories[Project][Code]

ActionRecognitionUsingaDistributedRepresentationofPoseandAppearance[Project]

RecognitionUsingRegions[Paper][Code]

2DArticulatedHumanPoseEstimation[Project]

FastHumanPoseEstimationUsingAppearanceandMotionviaMulti-DimensionalBoostingRegression[Paper][Code]

EstimatingHumanPosefromOccludedImages[Paper][Code]

Quasi-densewidebaselinematching[Project]

ChaLearnGestureChallenge:Principalmotion:PCA-basedreconstructionofmotionhistograms[Project]

RealTimeHeadPoseEstimationwithRandomRegressionForests[Project]

2DActionRecognitionServes3DHumanPoseEstimation[

AHoughTransform-BasedVotingFrameworkforActionRecognition[

MotionInterchangePatternsforActionRecognitioninUnconstrainedVideos[

2Darticulatedhumanposeestimationsoftware[Project]

Learninganddetectingshapemodels[code]

ProgressiveSearchSpaceReductionforHumanPoseEstimation[Project]

LearningNon-Rigid3DShapefrom2DMotion[Project]

十二、图像处理:

DistanceTransformsofSampledFunctions[Project]

TheComputerVisionHomepage[Project]

Efficientappearancedistancesbetweenwindows[code]

ImageExplorationalgorithm[code]

MotionMagnification运动放大[Project]

BilateralFilteringforGrayandColorImages双边滤波器[Project]

AFastApproximationoftheBilateralFilterusingaSignalProcessingApproach[

十三、一些实用工具:

EGT:aToolboxforMultipleViewGeometryandVisualServoing[Project][Code]

adevelopmentkitofmatlabmexfunctionsforOpenCVlibrary[Project]

FastArtificialNeuralNetworkLibrary[Project]

十四、人手及指尖检测与识别:

finger-detection-and-gesture-recognition[Code]

HandandFingerDetectionusingJavaCV[Project]

Handandfingersdetection[Code]

十五、场景解释:

NonparametricSceneParsingviaLabelTransfer[Project]

十六、光流Opticalflow:

Highaccuracyopticalflowusingatheoryforwarping[Project]

DenseTrajectoriesVideoDescription[Project]

SIFTFlow:DenseCorrespondenceacrossScenesanditsApplications[Project]

KLT:AnImplementationoftheKanade-Lucas-TomasiFeatureTracker[Project]

TrackingCarsUsingOpticalFlow[Project]

Secretsofopticalflowestimationandtheirprinciples[Project]

implmentationoftheBlackandAnandandenseopticalflowmethod[Project]

OpticalFlowComputation[Project]

BeyondPixels:ExploringNewRepresentationsandApplicationsforMotionAnalysis[Project]

ADatabaseandEvaluationMethodologyforOpticalFlow[Project]

opticalflowrelative[Project]

RobustOpticalFlowEstimation[Project]

opticalflow[Project]

十七、图像检索ImageRetrieval:

Semi-SupervisedDistanceMetricLearningforCollaborativeImageRetrieval[Paper][code]

十八、马尔科夫随机场MarkovRandomFields:

MarkovRandomFieldsforSuper-Resolution[Project]

AComparativeStudyofEnergyMinimizationMethodsforMarkovRandomFieldswithSmoothness-BasedPriors[Project]

十九、运动检测Motiondetection:

MovingObjectExtraction,UsingModelsorAnalysisofRegions[Project]

BackgroundSubtraction:ExperimentsandImprovementsforViBe[Project]

ASelf-OrganizingApproachtoBackgroundSubtractionforVisualSurveillanceApplications[Project]

:Anewchangedetectionbenchmarkdataset[Project]

ViBe-apowerfultechniqueforbackgrounddetectionandsubtractioninvideosequences[Project]

BackgroundSubtractionProgram[Project]

MotionDetectionAlgorithms[Project]

StuttgartArtificialBackgroundSubtractionDataset[Project]

ObjectDetection,MotionEstimation,andTracking[Project]

FeatureDetectionandDescription模型代码

GeneralLibraries:

VLFeat–Implementationofvariousfeaturedescriptors(includingSIFT,HOG,andLBP)andcovariantfeaturedetectors(includingDoG,Hessian,HarrisLaplace,HessianLaplace,MultiscaleHessian,MultiscaleHarris).:Software–S

OpenCV–Variousimplementationsofmodernfeaturedetectorsanddescriptors(SIFT,SURF,FAST,BRIEF,ORB,FREAK,etc.)

FastKeypointDetectorsforReal-timeApplications:

FAST–High-speedcornerdetectorimplementationforawidevarietyofplatforms

AGAST–(ECCV2010).

BinaryDescriptorsforReal-TimeApplications:

BRIEF–C++codeforafastandaccurateinterestpointdescriptor(notinvarianttorotationsandscale)(ECCV2010)

ORB–OpenCVimplementationoftheOriented-Brief(ORB)descriptor(invarianttorotations,butnotscale)

BRISK–(ICCV2011)

FREAK–FasterthanBRISK(invarianttorotationsandscale)(CVPR2012)

SIFTandSURFImplementations:

SIFT:VLFeat,OpenCV,OriginalcodebyDavidLowe,GPUimplementation,OpenSIFT

SURF:HerbertBay’scode,OpenCV,GPU-SURF

OtherLocalFeatureDetectorsandDescriptors:

VGGAffineCovariantfeatures–Oxfordcodeforvariousaffinecovariantfeaturedetectorsanddescriptors.

LIOPdescriptor–SourcecodefortheLocalIntensityorderPattern(LIOP)descriptor(ICCV2011).

LocalSymmetryFeatures–Sourcecodeformatchingoflocalsymmetryfeaturesunderlargevariationsinlighting,age,andreringstyle(CVPR2012).

GlobalImageDescriptors:

GIST–MatlabcodefortheGISTdescriptor

CENTRIST–Globalvisualdescriptorforscenecategorizationandobjectdetection(PAMI2011)

FeatureCodingandPooling

VGGFeatureEncodingToolkit–Sourcecodeforvariousstate-of-the-artfeatureencodingmethods–includingStandardhardencoding,Kernelcodebookencoding,Locality-constrainedlinearencoding,andFisherkernelencoding.

SpatialPyramidMatching–Sourcecodeforfeaturepoolingbasedonspatialpyramidmatching(widelyusedforimageclassification)

ConvolutionalNetsandDeepLearning

EBLearn–C++iersbasedonconvolutionalneuralnetworks.

Torch7–Providesamatlab-likeenvironmentforstate-of-the-artmachinelearningalgorithms,includingafastimplementationofconvolutionalneuralnetworks.

DeepLearning-Variouslinksfordeeplearningsoftware.

Part-BasedModels

DeformablePart-basedDetector–Libraryprovidedbytheauthorsoftheoriginalpaper(state-of-the-artinPASCALVOCdetectiontask)

EfficientDeformablePart-BasedDetector–Branch-and-Boundimplementationforadeformablepart-baseddetector.

AcceleratedDeformablePartModel–Efficientimplementationofamethodthatachievestheexactsameperformanceofdeformablepart-baseddetectorsbutwithsignificantacceleration(ECCV2012).

Coarse-to-FineDeformablePartModel–Fastapproachfordeformableobjectdetection(CVPR2011).

Poselets–C++andMatlabversionsforobjectdetectionbasedonposelets.

Part-basedFaceDetectorandPoseEstimation–Implementationofaunifiedapproachforfacedetection,poseestimation,andlandmarklocalization(CVPR2012).

AttributesandSemanticFeatures

RelativeAttributes–ModifiedimplementationofRankSVMtotrainRelativeAttributes(ICCV2011).

ObjectBank–Implementationofobjectbanksemanticfeatures(NIPS2010).SeealsoActionBank

Classemes,Picodes,andMeta-classfeatures–Softwareforextractinghigh-levelimagedescriptors(ECCV2010,NIPS2011,CVPR2012).

Large-ScaleLearning

AdditiveKernels–SourcecodeforfastadditivekernelSVMclassifiers(PAMI2013).

LIBLINEAR–Libraryforlarge-scalelinearSVMclassification.

VLFeat–ImplementationforPegasosSVMandHomogeneousKernelmap.

FastIndexingandImageRetrieval

FLANN–Libraryforperformingfastapproximatenearestneighbor.

KernelizedLSH–SourcecodeforKernelizedLocality-SensitiveHashing(ICCV2009).

ITQBinarycodes–CodeforgenerationofsmallbinarycodesusingIterativeQuantizationandotherbaselinessuchasLocality-Sensitive-Hashing(CVPR2011).

INRIAImageRetrieval–Efficientcodeforstate-of-the-artlarge-scaleimageretrieval(CVPR2011).

ObjectDetection

SeePart-basedModelsandConvolutionalNetsabove.

PedestrianDetectionat100fps–Veryfastandaccuratepedestriandetector(CVPR2012).

CaltechPedestrianDetectionBenchmark–Excellentresourceforpedestriandetection,withvariouslinksforstate-of-the-artimplementations.

OpenCV–EnhancedimplementationofViolaJonesreal-timeobjectdetector,withtrainedmodelsforfacedetection.

EfficientSubwindowSearch–Sourcecodeforbranch-and-boundoptimizationforefficientobjectlocalization(CVPR2008).

3DRecognition

Point-CloudLibrary–Libraryfor3Dimageandpointcloudprocessing.

ActionRecognition

ActionBank–SourcecodeforactionrecognitionbasedontheActionBankrepresentation(CVPR2012).

STIPFeatures–softwareforcomputingspace-timeinterestpointdescriptors

IndepentSubspaceAnalysis–LookforStackedISAforVideos(CVPR2011)

VelocityHistoriesofTrackedKeypoints-C++codeforactivityrecognitionusingthevelocityhistoriesoftrackedkeypoints(ICCV2009)

数据集

Attributes

AnimalswithAttributes–30,475imagesof50animalsclasseswith6pre-extractedfeaturerepresentationsforeachimage.

aYahooandaPascal–AttributeannotationsforimagescollectedfromYahooandPascalVOC2008.

FaceTracer–15,000facesannotatedwith10attributesandfiducialpoints.

PubFig–58,797faceimagesof200peoplewith73attributeclassifieroutputs.

[url=]LFW[/url]–13,233faceimagesof5,749peoplewith73attributeclassifieroutputs.

HumanAttributes–8,000

SUNAttributeDatabase–Large-scalesceneattributedatabasewithataxonomyof102attributes.

ImageNetAttributes–VarietyofattributelabelsfortheImageNetdataset.

Relativeattributes–

AttributeDiscoveryDataset–Imagesofshoppingcategoriesassociatedwithtextualdescriptions.

Fine-grainedVisualCategorization

Caltech-UCSDBirdsDataset–Hundredsofbirdcategorieswithannotatedpartsandattributes.

StanfordDogsDataset–20,000imagesof120breedsofdogsfromaroundtheworld.

Oxford-IIITPetDataset–37

LeedsButterflyDataset–832imagesof10speciesofbutterflies.

OxfordFlowerDataset–Hundredsofflowercategories.

FaceDetection

[url=]FDDB[/url]–UMassfacedetectiondatasetandbenchmark(5,000+faces)

CMU/MIT–Classicalfacedetectiondataset.

FaceRecognition

FaceRecognitionHomepage–Largecollectionoffacerecognitiondatasets.

[url=]LFW[/url]–UMassunconstrainedfacerecognitiondataset(13,000+faceimages).

NISTFaceHomepage–includesfacerecognitiongrandchallenge(FRGC),vortests(FRVT)andothers.

CMUMulti-PIE–containsmorethan750,000imagesof337people,with15differentviewsand19lightingconditions.

FERET–Classicalfacerecognitiondataset.

DengCai’sfacedatasetinMatlabFormat–EasytouseifyouwantplaywithsimplefacedatasetsincludingYale,ORL,PIE,andExtedYaleB.

SCFace–Low-resolutionfacedatasetcapturedfromsurveillancecameras.

HandwrittenDigits

MNIST–largedatasetcontainingatrainingsetof60,000examples,andatestsetof10,000examples.

PedestrianDetection

CaltechPedestrianDetectionBenchmark–10hoursofvideotakenfromavehicle,350

INRIAPersonDataset–Currentlyoneofthemostpopularpedestriandetectiondatasets.

ETHPedestrianDataset–Urbandatasetcapturedfromastereorigmountedonastroller.

TUD-BrusselsPedestrianDataset–Datasetwithimagepairsrecordedinancrowdedurbansettingwithanonboardcamera.

PASCALHumanDetection–Oneof20categoriesinPASCALVOCdetectionchallenges.

USCPedestrianDataset–Smalldatasetcapturedfromsurveillancecameras.

GenericObjectRecognition

ImageNet–Currentlythelargestvisualrecognitiondatasetintermsofnumberofcategoriesandimages.

TinyImages–80million32x32lowresolutionimages.

PascalVOC–Oneofthemostinfluentialvisualrecognitiondatasets.

Caltech101/Caltech256–Popularimagedatasetscontaining101and256objectcategories,respectively.

MITLabelMe–Onlineannotationtoolforbuildingcomputervisiondatabases.

SceneRecognition

MITSUNDataset–MITsceneunderstandingdataset.

UIUCFifteenSceneCategories–Datasetof15naturalscenecategories.

FeatureDetectionandDescription

VGGAffineDataset–Widelyuse

ActionRecognition

BenchmarkingActivityRecognition–CVPR2012tutorialcoveringvariousdatasetsforactionrecognition.

RGBDRecognition

RGB-DObjectDataset–Datasetcontaining300commonhouseholdobjects

算法代码

特征提取

SURF特征:(当然这只是其中之一)

LBP特征(一种纹理特征):

FastCornerDetection(OpenCV中的Fast算法):FASTCornerDetection--EdwardRosten

机器视觉

Asimpleobjectdetectorwithboosting(AwardedtheBestShortCoursePrizeatICCV2005,So了解adaboost的推荐之作):

Boosting(该网页上有相当全的Boosting的文章和几个Boosting代码,本人推荐):

AdaboostMatlab工具:

MultiBoost(不说啥了,多类Adaboost算法的程序):

TextonBoost(我们教研室王冠夫师兄的毕设):JamieShotton-Code

LibSvm的老爹(推荐):

ConditionalRandomFields(CRF论文+Code列表,推荐)

CRF++:YetAnotherCRFtoolkit

ConditionalRandomField(CRF)ToolboxforMatlab

TreeCRFs

LingPipe:Installation

HiddenMarkovModels(推荐)

隐马尔科夫模型(HiddenMarkovModels)系列之一-eaglex的专栏-博客频道-(推荐)

综合代码

CvPapers(好吧,牛吧网站,里面有ICCV,CVPR,ECCV,SIGGRAPH的论文收录,然后还有一些论文的代码搜集,要求加精!):

ComputerVisionSoftware(里面代码很多,并详细的给出了分类):

某人的WindowsLive(我看里面东东不少就收藏了):;id=3B6244088FD5A769!523

MATLABandOctaveFunctionsforComputerVisionandImageProcessing(这个里面的东西也很全,只是都是用Matlab和Octave开发的):

ComputerVisionResources(里面的视觉算法很多,给出了相应的论文和Code,挺好的):

MATLABFunctionsforMultipleViewGeometry(关于物体多视角计算的库):

EvolutiveAlgorithmbasedonNaïveBayesmodelsEstimation(单独列了一个算法的Code):

主页代码

PabloNegri'sHomePage

JianxinWu'shomepage

PeterCarbonetto

MarkovRandomFieldsforSuper-Resolution

DetectingandSketchingtheCommon

PedroFelzenszwalb

HaeJONG,SEO

CAP5416-ComputerVision

ParallelTrackingandMappingforSmallARWorkspaces(PTAM)

DevaRamanan-UCIrvine-ComputerVision

RaghuramanGopalan

HuiKong

JamieShotton-Post-DoctoralResearcherinComputerVision

Jean-YvesAUDIBERT

OlgaVeksler

StephenGould

Publications(LastUpdate:09/30/10)

KarimAli-FlowBoost

Asimplepartsandstructureobjectdetector

Code-OxfordBrookesVisionGroup

TakuKudo

行人检测

HistogramofOrientedGradient(Windows)

INRIAPedestriandetector

Poselets

WilliamRobsonSchwartz-Softwares

RPT@CVG

MainPage

SourceCode

PedestrianDetection

Class-SpecificHoughForestsforObjectDetection

JianxinWu'shomepage(就是上面的)

Berkeley大学做的PedestrianDetector,使用交叉核的支持向量机,特征使用HOG金字塔,提供Matlab和C++混编的代码:

视觉壁障

HighSpeedObstacleAvoidanceusingMonocularVisionandReinforcementLearning

TLD(2010年很火的tracking算法)

onlineboostingtrackers

BorisBabenko

OpticalFlowAlgorithmEvaluation(提供了一个动态贝叶斯网络框架,例如递归信息处理与分析、卡尔曼滤波、粒子滤波、序列蒙特卡罗方法等,C++写的)

物体检测算法

ObjectDetection

Softwareforobjectdetection

人脸检测

SourceCode

10个人脸检测项目

JianxinWu'shomepage(又是这货)

ICA独立成分分析

AnICApage-papers,code,demo,links(TonyBell)

FastICA

Cachedk-dtreesearchforICPalgorithms

滤波算法

卡尔曼滤波:TheKalmanFilter(终极网页)

BayesianFilteringLibrary:TheBayesianFilteringLibrary

路面识别

SourceCode

Vanishingpointdetectionforgeneralroaddetection

分割算法

MATLABNormalizedCutsSegmentationCode:software

超像素分割:SLICSuperpixels

Reference:

;tid=3126highlight=%BC%C6%CB%E3%BB%FA%CA%D3%BE%F5%B4%FA%C2%EB

免责声明:本文章如果文章侵权,请联系我们处理,本站仅提供信息存储空间服务如因作品内容、版权和其他问题请于本站联系