Computer Vision: Algorithms and Application書籍章節介紹及其源碼

ComputerVision:

Algorithms and Applications

Richard Szeliski

 本書網址:書的最好附錄中,我總結了一些對學生,教授和研究者有用的附加材料。這本書的網址http://szeliski.org/Book

C.1 數據集

一個關鍵就是用富有挑戰和典型的數據集來測試你算法的可靠性。當有背景或者他人的結果是可行的,這種測試可能甚至包含更多的信息(和質量更好)。

經過這些年,大量的數據集已經被提出來用於測試和評估計算機視覺算法。許多這些數據集和軟件被編入了計算機視覺的主頁。一些更新的網址,像CVonline

(http://homepages.inf.ed.ac.uk/rbf/CVonline ), VisionBib.Com (http://datasets.visionbib.com/ ), and Computer Vision online (http://computervisiononline.com/ ), 有更多最新的數據集和軟件。

下面,我列出了一些用的最多的數據集,我將它們讓章節排列以便它們聯繫更緊密。

 

第二章:圖像信息

CUReT: Columbia-Utrecht 反射率和紋理數據庫Reflectance and TextureDatabase,http://www1.cs.columbia.edu/CAVE/software/curet/  (Dana, van Ginneken, Nayaret al. 1999).

 

Middlebury Color Datasets:不同攝像機拍攝的圖像,註冊後用於研究不同的攝像機怎麼改變色域和彩色registeredcolor images taken by different cameras to study how they transform gamuts andcolors, http://vision.middlebury.edu/color/data/ Chakrabarti, Scharstein, and Zickler 2009).

 

第三章:圖像處理

Middlebury test datasets forevaluating MRF minimization/inference algorithms評估隱馬爾科夫隨機場最小化和推斷算法,

http://vision.middlebury.edu/MRF/results/ (Szeliski, Zabih, Scharstein et al. 2008).

 

第四章:特徵檢測和匹配

Affine Covariant Featuresdatabase(反射協變的特徵數據集) for evaluating feature detector and descriptor matching quality andrepeatability(評估特徵檢測和描述匹配的質量和定位精度),   http://www.robots.ox.ac.uk/~vgg/research/affine/

(Miko-lajczyk and Schmid 2005;Mikolajczyk, Tuytelaars, Schmid et al. 2005).

 

Database of matched imagepatches for learning (圖像斑塊匹配學習數據庫)and feature descriptor evaluation(特徵描述評估數據庫),

http://cvlab.epfl.ch/~brown/patchdata/patchdata.html

(Winder and Brown 2007;Hua,Brown, and Winder 2007).

 

第五章;分割

BerkeleySegmentation Dataset(分割數據庫) and Benchmark of 1000 images labeled by 30 humans,(30個人標記的1000副基準圖像)along with an evaluation, http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/segbench/  (Martin, Fowlkes, Tal et al.2001).

 

Weizmann segmentationevaluation database of 100 grayscale images with ground truth segmentations,

http://www.wisdom.weizmann.ac.il/~vision/Seg EvaluationDB/index.html

 (Alpert, Galun, Basri et al. 2007).

 

第八章:稠密運動估計

TheMiddlebury optic flow evaluation(光流評估) Web site, http://vision.middlebury.edu/flow/data/

(Baker,Scharstein, Lewis et al. 2009).

 

The Human-Assisted MotionAnnotation database,(人類輔助運動數據庫)

http://people.csail.mit.edu/celiu/motionAnnotation/  (Liu, Freeman, Adelson etal. 2008)

 

第十章:計算機攝像學

High DynamicRange radiance(輻射)maps, http://www.debevec.org/Research/HDR/

(De-bevecand Malik 1997).

 

Alpha matting evaluation Website, http://alphamatting.com/ (Rhemann, Rother, Wang

et al. 2009).

 

第十一章:Stereo correspondence立體對應

Middlebury Stereo Datasets andEvaluation, http://vision.middlebury.edu/stereo/  (Scharstein

and Szeliski 2002).

 

StereoClassification(立體分類) and Performance Evaluation(性能評估) of different aggregation(聚類) costs for stereo matching(立體匹配), http://www.vision.deis.unibo.it/spe/SPEHome.aspx  (Tombari, Mat-

toccia, Di Stefano et al.2008).

 

Middlebury Multi-View StereoDatasets,

http://vision.middlebury.edu/mview/data/  (Seitz,Curless, Diebel etal. 2006).

 

Multi-view and Oxford Collegesbuilding reconstructions,

http://www.robots.ox.ac.uk/~vgg/data/data-mview.html .

 

Multi-View Stereo Datasets, http://cvlab.epfl.ch/data/strechamvs/  (Strecha, Fransens,

and Van Gool 2006).

 

Multi-View Evaluation,  http://cvlab.epfl.ch/~strecha/multiview/ (Strecha, von Hansen,

Van Gool et al. 2008).

 

第十二章:3D重建

HumanEva: synchronized video(同步視頻) and motion capture (動作捕捉)dataset for evaluation ofarticulated human motion,http://vision.cs.brown.edu/humaneva/  Sigal, Balan, and Black 2010).

 

第十三章:圖像渲染

The (New) Stanford Light FieldArchive, http://lightfield.stanford.edu/ 

(Wilburn, Joshi,Vaish et al.2005).

 

Virtual Viewpoint Video:multi-viewpoint video with per-frame depth maps,

 http://research.microsoft.com/en-us/um/redmond/groups/ivm/vvv/  (Zitnick, Kang, Uytten-

daele et al. 2004).

 

第十四章:識別

查找一系列的視覺識別數據庫,在表14.1–14.2.除了那些,這裏還有:

Buffy pose classes, http://www.robots.ox.ac.uk/~vgg/data/  buffy pose classes/ andBuffy

stickmen V2.1, http://www.robots.ox.ac.uk/~vgg/data/stickmen/index.html  (Ferrari,Marin-

Jimenez, and Zisserman 2009;Eichner and Ferrari 2009).

 

H3D database of pose/jointannotated photographs of humans,

 http://www.eecs.berkeley.edu/~lbourdev/h3d/   (Bourdev and Malik 2009).

 

Action Recognition Datasets,http://www.cs.berkeley.edu/projects/vision/action, has point-

ers toseveral datasets for action and activity recognition, as well as some papers.(有一些關於人活動和運動的數據庫和論文) The humanaction database at http://www.nada.kth.se/cvap/actions/  包含更多的行動序列。

 

C.2 軟件資源

一個對於計算機視覺算法最好的資源就是開源視覺圖像庫(opencv)(http://opencv.willowgarage.com/wiki/),他有在intel的Gary Bradski和他的同事開發,現在由Willow Garage (Bradsky and Kaehler 2008)維護和擴展。一部分可利用的函數在http://opencv.willowgarage.com/documentation/cpp/中:

 圖像處理和變換 (濾波,形態學,金字塔);

 圖像幾何學的變換 (旋轉,改變大小);

混合圖像變換 (傅里葉變換,距離變換);

 直方圖;

 分割 (分水嶺, mean shift);

 特徵檢測 (Canny, Harris, Hough, MSER, SURF);

 運動分析和物體分析 (Lucas–Kanade, mean shift);

 相機矯正和3D重建

 機器學習 (k nearest neighbors, 支持向量機, 決策樹, boost-

ing, 隨機樹, expectation-maximization, 和神經網絡).

 

Intel的Performance Primitives (IPP)library, http://software.intel.com/en-us/intel-ipp/,包含

各種各樣的圖像處理任務的最佳優化代碼,許多opencv中的例子利用了這個庫,加入他安裝了,程序運行得更快。依據功能,他和Opencv有很多相同的運算處理,並且加上了額外的庫針對圖像視頻壓縮,信號語音處理和矩陣代數。

 

MTALAB中的Image Processing Toolbox圖像處理工具,http://www.mathworks.com/products/image/,包含常規的處理,空域變換(旋轉,改變大小),常規正交,圖像分析和統計學(變邊緣,哈弗變換),圖像增強(自適應直方圖均衡,中值濾波),圖像恢復(去模糊),線性濾波(卷積),圖像變換(傅里葉,離散餘弦變換)和形態學操作(連通域和距離變換)

 

兩個比較舊的庫,它們沒有被髮展,但是包含了一些的有用的常規操作:

VXL (C++Libraries for Computer Vision Research and Implemen-tation, http://vxl.sourceforge.net/)

LTI-Lib 2 (http://www.ie.itcr.ac.cr/palvarado/ltilib-2/homepage/ ).

 

 圖像編輯和視圖包,例如Windows Live Photo Gallery, iPhoto, Picasa,GIMP, 和 IrfanView,它們對執行這些處理非常有用:常規處理任務,格式轉換,觀測你的結果。它們同樣可以用於對圖像處理算法有趣的實現參考,例如色調調整和去噪。

 

這裏他也有一些軟件包和基礎框架對你建一個實時視頻處理的DEMOS很有用,Vision on Tap(http://www.visionontap.com/ )提供一個可以實時處理你的網絡攝像頭的網頁服務(Chiu and Raskar 2009)。Video-Man (VideoManager,http://videomanlib.sourceforge.net/處理實時的基於視頻的DEMOS和應用非常有用,你也可以用MATLAB中的imread直接從任何URl(例如網絡攝像頭)中讀取視頻。

下面,我列出了一些額外的網絡資源,讓章節排列以便它們看起來聯繫更緊密:

 

第三章:圖像處理

matlabPyrTools—MATLAB 下的源碼對於拉普拉斯變換,金字塔, QMF/小波, 和

steerable pyramids, http://www.cns.nyu.edu/~lcv/software.php  (Simoncelli and Adel-

son 1990a; Simoncelli,Freeman, Adelson et al. 1992).

 

BLS-GSM 圖像去噪, http://decsai.ugr.es/~javier/denoise/  (Portilla, Strela,Wain-

wright et al. 2003).

 

Fast bilateral filtering code(快速雙邊濾波), http://people.csail.mit.edu/jiawen/#code (Chen, Paris, and Durand 2007).

 

C++ implementation of the fastdistance transform algorithm,

http://people.cs.uchicago.edu/~pff/dt/  (Felzenszwalb andHuttenlocher 2004a).

 

GREYC’s Magic Image Converter,including image restoration software using regularization and anisotropicdiffusion,http://gmic.sourceforge.net/gimp.shtml (Tschumperl´ e and Deriche 2005).

 

第四章:圖像特徵檢測和匹配

VLFeat, 一個開放便捷的計算機視覺算法庫

http://vlfeat.org/ (Vedaldi and Fulkerson 2008).

 

SiftGPU: A GPU Implementationof Scale Invariant Feature Transform (SIFT),

GPU實現的尺度特徵性變換

http://www.cs.unc.edu/~ccwu/siftgpu/  (Wu 2010).

 

SURF: Speeded Up RobustFeatures, http://www.vision.ee.ethz.ch/~surf/

(Bay, Tuyte-laars, and VanGool 2006).

 

FAST corner detection, http://mi.eng.cam.ac.uk/~er258/work/fast.html

 (Rosten and Drum-mond 2005, 2006).

 

Linux binaries for affineregion detectors and descriptors, as well as MATLAB files to

compute repeatability andmatching scores,

http://www.robots.ox.ac.uk/~vgg/research/affine/

 

Kanade–Lucas–Tomasi featuretrackers: KLT, http://www.ces.clemson.edu/~stb/klt/ (Shi and Tomasi 1994);

GPU-KLT, http://cs.unc.edu/~cmzach/opensource.html  (Zach,Gallup, and Frahm2008); Lucas–Kanade 20 Years On,http://www.ri.cmu.edu/projects/project 515.html  (Baker and Matthews 2004).

第五章:分割

高效的基於圖形的分割http://people.cs.uchicago.edu/~pff/segment 

(Felzenszwalb and Huttenlocher2004b).

 

EDISON, 邊緣檢測和圖像追蹤,

http://coewww.rutgers.edu/riul/research/code/EDISON/

 (Meer and Georgescu 2001; Comaniciu and Meer2002).

 

Normalized cuts segmentationincluding intervening contours,

http://www.cis.upenn.edu/~jshi/software/

(Shi and Malik 2000; Malik,Belongie, Leung et al. 2001).

 

Segmentation by weightedaggregation (SWA),利用加權集合的分割

http://www.cs.weizmann.ac.il/~vision/SWA  (Alpert, Galun, Basri et al.2007).

 

第六章:基於特徵的對齊和校準

Non-iterative PnP algorithm,(非迭代PnP算法)

http://cvlab.epfl.ch/software/EPnP  (Moreno-Noguer, Lep-etit, and Fua 2007).

 

Tsai Camera Calibration(相機矯正) Software,

http://www-2.cs.cmu.edu/~rgw/TsaiCode.html  (Tsai 1987).

 

Easy CameraCalibration Toolkit,(簡易相機校準工具包) http://research.microsoft.com/en-us/um/people/zhang/ Calib/ (Zhang 2000).

 

Camera Calibration Toolbox forMATLAB,

http://www.vision.caltech.edu/bouguetj/calib doc/ ; a C version is included in OpenCV.

 

MATLAB functions for multipleview geometry,

http://www.robots.ox.ac.uk/~vgg/hzbook/code/  (Hartley and Zisserman2004).

 

第七章:運動重建

SBA: A generic sparse bundle(稀疏束) adjustment C/C++ package basedon the Levenberg–

Marquardt algorithm, http://www.ics.forth.gr/~lourakis/sba/  (Lourakis and Argyros 2009).

 

Simple sparse bundleadjustment (SSBA), http://cs.unc.edu/~cmzach/opensource.html .

 

Bundler, structure from motionfor unordered image collections(無序圖像集),

http://phototour.cs.washington.edu/bundler/   (Snavely, Seitz, and Szeliski 2006).

 

第八章:稠密運動估計

光流, http://www.cs.brown.edu/~black/code.html (Black and Anan-

dan 1996).

 

Optical flow(光流) using total variation(全變量差) and conjugate gradientdescent(共軛梯度下降),http://people.csail.mit.edu/celiu/OpticalFlow/  (Liu 2009).

 

TV-L1 optical flow on the GPU, http://cs.unc.edu/~cmzach/opensource.html

(Zach,Pock, and Bischof2007a).

 

elastix: atoolbox for rigid(剛性) and nonrigid(非剛性) registration of images(配準圖像), http://elastix.isi.uu.nl/ (Klein, Staring, and Pluim 2007).

 

Deformable image registration(可變形的配準圖像) using discreteoptimization(離散最優化), http://www.mrf-registration.net/deformable/index.html

 (Glocker, Komodakis, Tziritas et al. 2008).

 

第九章:圖像縫合

Microsoft Research ImageCompositing Editor for stitching images,(圖像拼接,圖像合成)

http://research.microsoft.com/en-us/um/redmond/groups/ivm/ice/ .

 

第十章:計算機攝影學

HDRShop software for combiningbracketed exposures(包圍式曝光) into high-dynamic range radiance images,http://projects.ict.usc.edu/graphics/HDRShop/.

 

Super-resolution(超分辨率) code,

http://www.robots.ox.ac.uk/~vgg/software/SR/  (Pickup 2007;Pickup, Capel,Roberts et al. 2007, 2009).

 

第十一章:立體對應

StereoMatcher, standalone C++stereo matching code,

http://vision.middlebury.edu/stereo/code/  (Scharstein and Szeliski2002).

 

Patch-based multi-view stereosoftware (PMVS Version 2),

http://grail.cs.washington.edu/software/pmvs/  (Furukawa and Ponce 2011).

 

第十二章:3D重建

Scanalyze: a system foraligning and merging range data,

http://graphics.stanford.edu/software/scanalyze/  (Curless and Levoy 1996).

 

MeshLab: software forprocessing, editing, and visualizing unstructured 3D triangular

meshes, http://meshlab.sourceforge.net/.

 

VRML viewers (various) arealso a good way to visualize texture-mapped 3D models.

 

節 12.6.4: Whole body modeling andtracking(全身建模和追蹤)

 

Bayesian 3D person tracking(貝葉斯3D人體追蹤), http://www.cs.brown.edu/~black/code.html  (Sidenbladh,Black, and Fleet2000; Sidenbladh and Black 2003).

 

HumanEva: baseline code forthe tracking of articulated human motion,

http://vision.cs.brown.edu/humaneva/   (Sigal, Balan, and Black 2010).

 

節 14.1.1: Face detection(人臉檢測)

 

Sample face detection code andevaluation tools,

http://vision.ai.uiuc.edu/mhyang/face-detection-survey.html.

 

節 14.1.2: Pedestrian detection(行人追蹤)

 

A simple object detector withboosting,

http://people.csail.mit.edu/torralba/shortCourseRLOC/boosting/boosting.html

 (Hastie, Tibshirani, and Friedman 2001;Torralba, Murphy, and Freeman 2007).

 

Discriminatively(有區別) trained deformable(可變形) part models, http://people.cs.uchicago.edu/~pff/latent/  (Felzenszwalb, Girshick,McAllester et al. 2010).

 

Upper-body detector(上身檢測),

http://www.robots.ox.ac.uk/~vgg/software/UpperBody/  (Ferrari,Marin-Jimenez, andZisserman 2008).

 

2D articulated human poseestimation software,

http://www.vision.ee.ethz.ch/~calvin/articulated_human_pose_estimation_code/  (Eichner and Ferrari 2009).

 

節 14.2.2: Active appearance and 3Dshape models

 

AAMtools: An active appearancemodeling toolbox,

 http://cvsp.cs.ntua.gr/software/AAMtools/  (Papandreou and Maragos2008).

 

節 14.3: Instance recognition

 

FASTANN and FASTCLUSTER forapproximate k-means (AKM),

http://www.robots.ox.ac.uk/~vgg/software/ (Philbin, Chum, Isard et al. 2007).

 

Feature matching using fastapproximate nearest neighbors,

http://people.cs.ubc.ca/~mariusm/index.php/FLANN/FLANN  (Muja and Lowe 2009).

 

 

節 14.4.1: Bag of words(詞袋)

 

Two bag of words classifiers, http://people.csail.mit.edu/fergus/iccv2005/bagwords.html

(Fei-Fei and Perona 2005;Sivic, Russell, Efros et al. 2005).

 

Bag of features andhierarchical(分層) k-means, http://www.vlfeat.org/  (Nist´ er and Stew´enius2006; Nowak, Jurie, and Triggs 2006).

 

節 14.4.2: Part-based models

 

A simple parts and structureobject detector,

http://people.csail.mit.edu/fergus/iccv2005/partsstructure.html

 (Fischler and Elschlager 1973; Felzenszwalband Huttenlocher 2005).

 

節 14.5.1: Machine learning software

 

Support vector machines (SVM)software (

http://www.support-vector-machines.org/SVM soft.html )

 包含很多支持向量機的庫,

SVMlight http://svmlight.joachims.org/ ;

 LIBSVM, http://www.csie.ntu.edu.tw/~cjlin/libsvm/(Fan, Chen,and Lin 2005);

LIBLINEAR, http://www.csie.ntu.edu.tw/~cjlin/liblinear/  (Fan,Chang, Hsieh et al.2008).

 

Kernel Machines: links to SVM,Gaussian processes, boosting, and other machine

learning algorithms, http://www.kernel-machines.org/software .

 

Multiple kernels for imageclassification,

http://www.robots.ox.ac.uk/~vgg/software/MKL 

 (Varma and Ray 2007; Vedaldi, Gulshan, Varmaet al. 2009).

 

附錄 A.1–A.2: Matrix decompositions(矩陣分解) and linear least squares(線性最小乘)

 

BLAS (BasicLinear Algebra Subprograms基本線性代數子程序),

http://www.netlib.org/blas/  (Blackford,Demmel, Dongarraet al. 2002).

 

LAPACK (Linear Algebra(線性代數) PACKage),

http://www.netlib.org/lapack/  (Anderson, Bai,Bischof etal. 1999).

 

GotoBLAS, http://www.tacc.utexas.edu/tacc-projects/.

 

ATLAS (Automatically TunedLinear Algebra Software),

http://math-atlas.sourceforge.net/  (Demmel, Dongarra, Eijkhoutet al. 2005).

 

Intel Math Kernel Library(MKL), http://software.intel.com/en-us/intel-mkl/.

 

AMD CoreMath Library (ACML),

http://developer.amd.com/cpu/Libraries/acml/Pages/default.aspx .

 

Robust PCA code(魯棒主成分分析), http://www.salle.url.edu/~ftorre/papers/rpca2.html

 (De la Torre and Black 2003).

 

Appendix A.3: Non-linear leastsquares非線性最小二乘

 

MINPACK, http://www.netlib.org/minpack/.

 

levmar: Levenberg–Marquardtnonlinear least squares algorithms, 非線性最小二乘

http://www.ics.forth.gr/~lourakis/levmar/  (Madsen, Nielsen, andTingleff 2004).

 

附錄 A.4–A.5: Direct(直接) and iterative(迭代) sparse matrix(稀疏矩陣) solvers

 

SuiteSparse (variousreordering algorithms, 各種各樣的重排算法CHOLMOD) and SuiteSparse QR,http://www.cise.ufl.edu/research/sparse/SuiteSparse/  (Davis 2006, 2008).

 

PARDISO (iterative and sparsedirect solution),  http://www.pardiso-project.org/.

 

TAUCS (sparse direct,iterative, out of core, preconditioners),

http://www.tau.ac.il/~stoledo/taucs/ .

 

HSL Mathematical SoftwareLibrary,  http://www.hsl.rl.ac.uk/index.html .

 

Templatesfor the solution of linear systems(線性系統解決問題的模板), http://www.netlib.org/linalg/html templates/Templates.html (Barrett, Berry, Chan et al.1994). Download the PDF for instructions(說明) on how to get the software.

 

ITSOL,MIQR, and other sparsesolvers,

 http://www-users.cs.umn.edu/~saad/software/  (Saad 2003).

 

ILUPACK, http://www-public.tu-bs.de/~bolle/ilupack/ .

 

附錄 B: Bayesian modeling and inference(貝葉斯建模和推斷)

 

Middleburysource code for MRF minimization(隱馬爾科夫隨機場最小化), http://vision.middlebury.edu/MRF/code/  (Szeliski, Zabih, Scharsteinet al. 2008).

 

C++ code for efficient beliefpropagation for early vision,

http://people.cs.uchicago.edu/~pff/bp/  (Felzenszwalb andHuttenlocher 2006).

 

FastPD MRF optimization(最優化) code,

http://www.csd.uoc.gr/~komod/FastPD  (Komodakisand Tziritas2007a; Komodakis, Tziritas, and Paragios 2008)

 

 

算法 C.1   Calgorithm for Gaussian random noise generation, using the Box–Mullertransform.

C描述的利用Box–Muller 變換產生高斯隨機噪聲

double urand()

{

return ((double)rand()) / ((double) RAND MAX);

}

void grand(double& g1, double& g2)

{

#ifndef M_PI

#define M_PI 3.14159265358979323846

#endif // M_PI

double n1 = urand();

double n2 = urand();

double x1 = n1 + (n1 == 0); /* guardagainst log(0) */

double sqlogn1 = sqrt(-2.0 * log (x1));

double angl = (2.0 * M PI) * n2;

g1 = sqlogn1 * cos(angl);

g2 = sqlogn1 * sin(angl);

}

 

 

高斯噪聲的產生。許多基本的軟件包產生一些不同的隨機的噪聲(例如 運行在unix上的rand()),但是並不是所有的都有高斯隨機噪聲發生器。計算一個離散隨機常量,你可以用Box–Mullertransform (Box and Muller 1958),他的c代碼在算法C.1中給出了,注意這個運行結果是返回一對隨機變量。相關的產生高斯隨機變量的方由Thomas, Luk, Leong et al. (2007)提出。

 

僞彩色產生。在很多應用中,很方便給圖像加上標記(或者給圖像特徵比如線)。一個最簡單的方式就是給不同的標記不同的顏色。在我的工作中,我發現用RGB立體色彩系給不同的標記賦予標準均勻的色彩是很方便的。

對於每一個(非消極)標記值,considerthe bits as being split among the three color channel,例如對於一個比特值爲9的值,

這個值可以被標記爲RGBRGBRGB,獲得三基色中的每一種顏色值後,顛倒比特值,結果是低位的比特值變化的最快。

實際上,對於一個八比特的顏色通道,這個比特值的顛倒可以被存在一個表或者一個存儲提前計算好的記錄有由標記值向僞彩色的改變的完整表。

圖 8.16 顯示了這樣一個僞彩色繪製的例子.

 

GPU實現

 

GPU的出現,可以處理像素着色和計算着色,導致了實時應用的快速計算機視覺算法的發展,例如,分割,追蹤,立體和運動估計((Pock, Unger, Cremerset al. 2008; Vineet and Narayanan 2008; Zach,Gallup, and Frahm 2008)。一個好的資源來學習這些算法就是CVPR 2008 上關於Visual Computer Visionon GPUs的workshop。

http://www.cs.unc.edu/~jmf/Workshop_on_Computer_Vision_on_GPU.html他的論文可以在CVPR2008的會議集的DVD中找到。額外的關於GPU算法資源包括GPGPU網址和小組討論http://gpgpu.org/還有OpenVIDIAWeb site,http://openvidia.sourceforge.net/index.php/OpenVIDIA

 

C.3 PPT和講稿

正如我在前言中提到的,我希望提供和書中材料相一致的PPT,直到這些全部準備好,你最好的方式去看我在華盛頓大學上課時的PPT,和一寫相關課程中用到的教案。

這裏是一些這樣的課程列表:

UW 455:Undergraduate Computer Vision,

 http://www.cs.washington.edu/education/courses/455/.

 

UW576:Graduate Computer Vision,

 http://www.cs.washington.edu/education/courses/576.

 

StanfordCS233B: Introduction to Computer Vision,

http://vision.stanford.edu/teaching/cs223b/.

 

MIT6.869: Advances in Computer Vision,

 http://people.csail.mit.edu/torralba/courses/6.869/6.869.computervision.htm.

 

Berkeley CS 280: Computer Vision, http://www.eecs.berkeley.edu/~trevor/CS280.html

 

UNC COMP776: Computer Vision, http://www.cs.unc.edu/~lazebnik/spring10.

 

Middlebury CS 453: Computer Vision,

 http://www.cs.middlebury.edu/~schar/courses/cs453-s10/.

 

Related courses have also been taught onthe topic of Computational Photography, e.g.,

 

CMU 15-463: Computational Photography, http://graphics.cs.cmu.edu/courses/15-463/.

 

MIT 6.815/6.865: Advanced ComputationalPhotography,

http://stellar.mit.edu/S/course/6/sp09/6.815

 

Stanford CS 448A: Computational photographyon cell phones,

http://graphics.stanford.edu/courses/cs448a-10/.

 

SIGGRAPH courses on ComputationalPhotography,

http://web.media.mit.edu/~raskar/photo/.

 

這裏還有一些最好的關於各種計算機視覺主題的在線講稿,例如:belief propagation and graph cuts,它們在UW-MSR Course of Vision Algo-rithms http://www.cs.washington.edu/education/courses/577/04sp/

 

C.4 參考文獻:

這本的所有參考文獻在這本書的網站上,一個幾乎所有的計算機視覺的出版物都引用的更全面的部分註解書目由Keith Price維http://iris.usc.edu/Vision-Notes/bibliography/contents.html.

這裏還有一個可搜索的計算機圖形學的參考書目http://www.siggraph.org/publications/bibliography/另外技術論文比較好的資源是GoogleScholar 和 CiteSeerX。



原文地址:

http://blog.csdn.net/senlerliu/article/details/49822093

微信掃碼關注我們:跟着數理化走天下


獲得更多的信息哦,一起交流,一起成長哦:微信號:跟着數理化走天下,純屬個人的交流,無盈利目的


發表評論
所有評論
還沒有人評論,想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.
相關文章