机器视觉分析源码列表

转至：http://www.yuanyong.org

一、特征提取Feature Extraction：

·             SIFT [1] [Demo program][SIFT Library] [VLFeat]

·             PCA-SIFT [2] [Project]

·             Affine-SIFT [3] [Project]

·             SURF [4] [OpenSURF] [Matlab Wrapper]

·             Affine Covariant Features[5] [Oxford project]

·             MSER [6] [Oxford project] [VLFeat]

·             Geometric Blur [7] [Code]

·             Local Self-similarityDescriptor [8] [Oxford implementation]

·             Global and EfficientSelf-similarity [9] [Code]

·             Histogram of OrientedGraidents [10] [INRIA Object Localization Toolkit] [OLT toolkit for Windows]

·             GIST [11] [Project]

·             Shape Context [12] [Project]

·             Color Descriptor [13] [Project]

·             Pyramids of Histograms ofOriented Gradients [Code]

·             Space-Time InterestPoints (STIP) [14][Project] [Code]

·             Boundary Preserving DenseLocal Regions [15][Project]

·             Weighted Histogram[Code]

·             Histogram-based InterestPoints Detectors[Paper][Code]

·             An OpenCV - C++implementation of Local Self similarity Descriptors [Project]

·             Fast SparseRepresentation with Prototypes[Project]

·             Corner Detection [Project]

·             AGAST Corner Detector:faster than FAST and even FAST-ER[Project]

·             Real-time Facial FeatureDetection using Conditional Regression Forests[Project]

·             Global and EfficientSelf-similarity for Object Classification and Detection[code]

·             WαSH: Weighted α-Shapesfor Local Feature Detection[Project]

·             HOG[Project]

·             Online Selection ofDiscriminative Tracking Features[Project]

 

二、图像分割Image Segmentation：

·             Normalized Cut [1] [Matlab code]

·             Gerg Mori’ Superpixelcode [2] [Matlab code]

·             Efficient Graph-basedImage Segmentation [3] [C++ code] [Matlab wrapper]

·             Mean-Shift ImageSegmentation [4] [EDISON C++ code] [Matlab wrapper]

·             OWT-UCM HierarchicalSegmentation [5] [Resources]

·             Turbepixels [6] [Matlab code 32bit] [Matlab code64bit] [Updated code]

·             Quick-Shift [7] [VLFeat]

·             SLIC Superpixels [8] [Project]

·             Segmentation by MinimumCode Length [9] [Project]

·             Biased Normalized Cut[10] [Project]

·             Segmentation Tree [11-12][Project]

·             Entropy Rate SuperpixelSegmentation [13] [Code]

·             Fast Approximate EnergyMinimization via Graph Cuts[Paper][Code]

·             Efficient Planar GraphCuts with Applications in Computer Vision[Paper][Code]

·             Isoperimetric GraphPartitioning for Image Segmentation[Paper][Code]

·             Random Walks for ImageSegmentation[Paper][Code]

·             Blossom V: A newimplementation of a minimum cost perfect matching algorithm[Code]

·             An Experimental Comparisonof Min-Cut/Max-Flow Algorithms for Energy Minimization in Computer Vision[Paper][Code]

·             Geodesic Star Convexityfor Interactive Image Segmentation[Project]

·             Contour Detection andImage Segmentation Resources[Project][Code]

·             Biased Normalized Cuts[Project]

·             Max-flow/min-cut[Project]

·             Chan-Vese Segmentationusing Level Set[Project]

·             A Toolbox of Level SetMethods[Project]

·             Re-initialization FreeLevel Set Evolution via Reaction Diffusion[Project]

·             Improved C-V activecontour model[Paper][Code]

·             A Variational MultiphaseLevel Set Approach to simultaneous Segmentation and Bias Correction[Paper][Code]

·             Level Set Method Researchby Chunming Li[Project]

·             ClassCut for UnsupervisedClass Segmentation[code]

·             SEEDS: SuperpixelsExtracted via Energy-Driven Sampling [Project][other]

 

三、目标检测Object Detection：

·             A simple object detectorwith boosting [Project]

·             INRIA Object Detectionand Localization Toolkit [1] [Project]

·             Discriminatively TrainedDeformable Part Models [2] [Project]

·             Cascade Object Detectionwith Deformable Part Models [3] [Project]

·             Poselet [4] [Project]

·             Implicit Shape Model [5][Project]

·             Viola and Jones’s FaceDetection [6] [Project]

·             Bayesian Modelling ofDyanmic Scenes for Object Detection[Paper][Code]

·             Hand detection usingmultiple proposals[Project]

·             Color Constancy, IntrinsicImages, and Shape Estimation[Paper][Code]

·             Discriminatively traineddeformable part models[Project]

·             Gradient Response Mapsfor Real-Time Detection of Texture-Less Objects: LineMOD [Project]

·             Image Processing On Line[Project]

·             Robust Optical FlowEstimation[Project]

·             Where's Waldo: MatchingPeople in Images of Crowds[Project]

·             Scalable Multi-classObject Detection[Project]

·             Class-Specific HoughForests for Object Detection[Project]

·             Deformed LatticeDetection In Real-World Images[Project]

·             Discriminatively traineddeformable part models[Project]

 

四、显著性检测Saliency Detection：

·             Itti, Koch, and Niebur’saliency detection [1] [Matlab code]

·             Frequency-tuned salientregion detection [2] [Project]

·             Saliency detection usingmaximum symmetric surround [3] [Project]

·             Attention via InformationMaximization [4] [Matlab code]

·             Context-aware saliencydetection [5] [Matlab code]

·             Graph-based visualsaliency [6] [Matlab code]

·             Saliency detection: Aspectral residual approach. [7] [Matlab code]

·             Segmenting salientobjects from images and videos. [8] [Matlab code]

·             Saliency Using Naturalstatistics. [9] [Matlab code]

·             Discriminant Saliency forVisual Recognition from Cluttered Scenes. [10] [Code]

·             Learning to Predict WhereHumans Look [11] [Project]

·             Global Contrast basedSalient Region Detection [12] [Project]

·             Bayesian Saliency via Lowand Mid Level Cues[Project]

·             Top-Down Visual Saliencyvia Joint CRF and Dictionary Learning[Paper][Code]

·             Saliency Detection: ASpectral Residual Approach[Code]

 

五、图像分类、聚类Image Classification, Clustering

·             Pyramid Match [1] [Project]

·             Spatial Pyramid Matching[2] [Code]

·             Locality-constrainedLinear Coding [3] [Project] [Matlab code]

·             Sparse Coding [4] [Project] [Matlab code]

·             Texture Classification[5] [Project]

·             Multiple Kernels forImage Classification [6] [Project]

·             Feature Combination [7] [Project]

·             SuperParsing [Code]

·             Large Scale CorrelationClustering Optimization[Matlab code]

·             Detecting and Sketchingthe Common[Project]

·             Self-Tuning SpectralClustering[Project][Code]

·             User Assisted Separationof Reflections from a Single Image Using a Sparsity Prior[Paper][Code]

·             Filters for TextureClassification[Project]

·             Multiple Kernel Learningfor Image Classification[Project]

·             SLIC Superpixels[Project]

 

六、抠图Image Matting

·             A Closed Form Solution toNatural Image Matting [Code]

·             Spectral Matting [Project]

·             Learning-based Matting [Code]

 

七、目标跟踪Object Tracking：

·             A Forest of Sensors- Tracking Adaptive Background Mixture Models [Project]

·             Object Tracking viaPartial Least Squares Analysis[Paper][Code]

·             Robust Object Trackingwith Online Multiple Instance Learning[Paper][Code]

·             Online Visual Trackingwith Histograms and Articulating Blocks[Project]

·             Incremental Learning forRobust Visual Tracking[Project]

·             Real-time CompressiveTracking[Project]

·             Robust Object Trackingvia Sparsity-based Collaborative Model[Project]

·             Visual Tracking viaAdaptive Structural Local Sparse Appearance Model[Project]

·             Online DiscriminativeObject Tracking with Local Sparse Representation[Paper][Code]

·             Superpixel Tracking[Project]

·             Learning HierarchicalImage Representation with Sparsity, Saliency and Locality[Paper][Code]

·             Online Multiple SupportInstance Tracking [Paper][Code]

·             Visual Tracking withOnline Multiple Instance Learning[Project]

·             Object detection andrecognition[Project]

·             Compressive SensingResources[Project]

·             Robust Real-Time VisualTracking using Pixel-Wise Posteriors[Project]

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

·             the HandVu：vision-based hand gestureinterface[Project]

·             Learning ProbabilisticNon-Linear Latent Variable Models for Tracking Complex Activities[Project]

 

八、Kinect：

·             Kinect toolbox[Project]

·             OpenNI[Project]

·             zouxy09 CSDN Blog[Resource]

·             FingerTracker手指跟踪[code]

 

九、3D相关：

·             3D Reconstruction of aMoving Object[Paper] [Code]

·             Shape From Shading UsingLinear Approximation[Code]

·             Combining Shape fromShading and Stereo Depth Maps[Project][Code]

·             Shape from Shading: ASurvey[Paper][Code]

·             A Spatio-TemporalDescriptor based on 3D Gradients (HOG3D)[Project][Code]

·             Multi-camera SceneReconstruction via Graph Cuts[Paper][Code]

·             A Fast MarchingFormulation of Perspective Shape from Shading under Frontal Illumination[Paper][Code]

·             Reconstruction:3D Shape,Illumination, Shading, Reflectance, Texture[Project]

·             Monocular Tracking of 3DHuman Motion with a Coordinated Mixture of Factor Analyzers[Code]

·             Learning 3-D SceneStructure from a Single Still Image[Project]

 

十、机器学习算法：

·             Matlab class forcomputing Approximate Nearest Nieghbor (ANN) [Matlab class providinginterface toANN library]

·             Random Sampling[code]

·             Probabilistic LatentSemantic Analysis (pLSA)[Code]

·             FASTANN and FASTCLUSTERfor approximate k-means (AKM)[Project]

·             Fast Intersection /Additive Kernel SVMs[Project]

·             SVM[Code]

·             Ensemble learning[Project]

·             Deep Learning[Net]

·             Deep Learning Methods forVision[Project]

·             Neural Network for Recognitionof Handwritten Digits[Project]

·             Training a deepautoencoder or a classifier on MNIST digits[Project]

·             THE MNIST DATABASE ofhandwritten digits[Project]

·             Ersatz：deep neural networks inthe cloud[Project]

·             Deep Learning [Project]

·             sparseLM : SparseLevenberg-Marquardt nonlinear least squares in C/C++[Project]

·             Weka 3: Data MiningSoftware in Java[Project]

·             Invited talk "ATutorial on Deep Learning" by Dr. Kai Yu (余凯)[Video]

·             CNN - Convolutionalneural network class[Matlab Tool]

·             Yann LeCun'sPublications[Wedsite]

·             LeNet-5, convolutionalneural networks[Project]

·             Training a deep autoencoderor a classifier on MNIST digits[Project]

·             Deep Learning大牛Geoffrey E. Hinton'sHomePage[Website]

·             Multiple InstanceLogistic Discriminant-based Metric Learning (MildML) and LogisticDiscriminant-based Metric Learning (LDML)[Code]

·             Sparse coding simulationsoftware[Project]

·             Visual Recognition andMachine Learning Summer School[Software]

 

十一、目标、行为识别Object, Action Recognition：

·             Action Recognition byDense Trajectories[Project][Code]

·             Action Recognition Usinga Distributed Representation of Pose and Appearance[Project]

·             Recognition UsingRegions[Paper][Code]

·             2D Articulated Human PoseEstimation[Project]

·             Fast Human PoseEstimation Using Appearance and Motion via Multi-Dimensional BoostingRegression[Paper][Code]

·             Estimating Human Posefrom Occluded Images[Paper][Code]

·             Quasi-dense wide baselinematching[Project]

·             ChaLearn GestureChallenge: Principal motion: PCA-based reconstruction of motion histograms[Project]

·             Real Time Head PoseEstimation with Random Regression Forests[Project]

·             2D Action RecognitionServes 3D Human Pose Estimation[

·             A Hough Transform-BasedVoting Framework for Action Recognition[

·             Motion InterchangePatterns for Action Recognition in Unconstrained Videos[

·             2D articulated human poseestimation software[Project]

·             Learning and detectingshape models [code]

·             Progressive Search SpaceReduction for Human Pose Estimation[Project]

·             Learning Non-Rigid 3DShape from 2D Motion[Project]

 

十二、图像处理：

·             Distance Transforms ofSampled Functions[Project]

·             The Computer VisionHomepage[Project]

·             Efficient appearancedistances between windows[code]

·             Image Explorationalgorithm[code]

·             Motion Magnification运动放大 [Project]

·             Bilateral Filtering forGray and Color Images双边滤波器 [Project]

·             A Fast Approximation ofthe Bilateral Filter using a Signal Processing Approach [

 

十三、一些实用工具：

·             EGT: a Toolbox forMultiple View Geometry and Visual Servoing[Project] [Code]

·             a development kit ofmatlab mex functions for OpenCV library[Project]

·             Fast Artificial NeuralNetwork Library[Project]

 

十四、人手及指纹检测与识别：

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

·             Hand and Finger Detectionusing JavaCV[Project]

·             Hand and fingersdetection[Code]

 

十五、场景解释：

·             Nonparametric SceneParsing via Label Transfer [Project]

 

十六、光流Optical flow：

·             High accuracy opticalflow using a theory for warping [Project]

·             Dense Trajectories VideoDescription [Project]

·             SIFT Flow: DenseCorrespondence across Scenes and its Applications[Project]

·             KLT: An Implementation ofthe Kanade-Lucas-Tomasi Feature Tracker [Project]

·             Tracking Cars UsingOptical Flow[Project]

·             Secrets of optical flowestimation and their principles[Project]

·             implmentation of theBlack and Anandan dense optical flow method[Project]

·             Optical Flow Computation[Project]

·             Beyond Pixels: ExploringNew Representations and Applications for Motion Analysis[Project]

·             A Database and EvaluationMethodology for Optical Flow[Project]

·             optical flow relative[Project]

·             Robust Optical FlowEstimation [Project]

·             optical flow[Project]

 

十七、图像检索Image Retrieval：

·             Semi-Supervised DistanceMetric Learning for Collaborative Image Retrieval [Paper][code]

 

十八、马尔科夫随机场Markov Random Fields：

·             Markov Random Fields forSuper-Resolution [Project]

·             A Comparative Study ofEnergy Minimization Methods for Markov Random Fields with Smoothness-BasedPriors [Project]

 

十九、运动检测Motion detection：

·             Moving Object Extraction,Using Models or Analysis of Regions [Project]

·             Background Subtraction:Experiments and Improvements for ViBe [Project]

·             A Self-OrganizingApproach to Background Subtraction for Visual Surveillance Applications [Project]

·             changedetection.net: Anew change detection benchmark dataset[Project]

·             ViBe - a powerfultechnique for background detection and subtraction in video sequences[Project]

·             Background SubtractionProgram[Project]

·             Motion DetectionAlgorithms[Project]

·             Stuttgart Artificial Background Subtraction Dataset[Project]

·             Object Detection, MotionEstimation, and Tracking[Project]

 

Feature Detection and Description

General Libraries:

·    VLFeat – Implementation ofvarious feature descriptors (including SIFT, HOG, and LBP) and covariantfeature detectors (including DoG, Hessian, Harris Laplace, Hessian Laplace,Multiscale Hessian, Multiscale Harris). Easy-to-use Matlab interface. See Modern features: Software – Slides providing a demonstration of VLFeat and also links to othersoftware. Check also VLFeat hands-onsession training

·    OpenCV – Various implementations of modern featuredetectors and descriptors (SIFT, SURF, FAST, BRIEF, ORB, FREAK, etc.)

 

Fast Keypoint Detectors for Real-timeApplications:

·    FAST – High-speed corner detector implementation for awide variety of platforms

·    AGAST – Even faster than the FAST corner detector. Amulti-scale version of this method is used for the BRISK descriptor (ECCV2010).

 

Binary Descriptors for Real-TimeApplications:

·    BRIEF – C++ code for a fast and accurate interest pointdescriptor (not invariant to rotations and scale) (ECCV 2010)

·    ORB – OpenCVimplementation of the Oriented-Brief (ORB) descriptor (invariant to rotations,but not scale)

·    BRISK – Efficient Binary descriptor invariant torotations and scale. It includes a Matlab mex interface. (ICCV 2011)

·    FREAK – Faster than BRISK(invariant to rotations and scale) (CVPR 2012)

 

SIFT and SURF Implementations:

·    SIFT: VLFeat, OpenCV, Original code by David Lowe, GPU implementation, OpenSIFT

·    SURF: Herbert Bay’scode, OpenCV, GPU-SURF

 

Other Local Feature Detectors and Descriptors:

·    VGG AffineCovariant features – Oxford code for variousaffine covariant feature detectors and descriptors.

·    LIOP descriptor – Sourcecode for the Local Intensity order Pattern (LIOP) descriptor (ICCV 2011).

·    Local SymmetryFeatures – Source code formatching of local symmetry features under large variations in lighting, age,and rendering style (CVPR 2012).

 

Global Image Descriptors:

·    GIST – Matlab code forthe GIST descriptor

·    CENTRIST – Global visual descriptor for scene categorizationand object detection (PAMI 2011)

 

Feature Coding and Pooling

·    VGG FeatureEncoding Toolkit – Sourcecode for various state-of-the-art feature encoding methods – including Standardhard encoding, Kernel codebook encoding, Locality-constrained linear encoding,and Fisher kernel encoding.

·    Spatial PyramidMatching – Source code forfeature pooling based on spatial pyramid matching (widely used for imageclassification)

 

Convolutional Nets and Deep Learning

·    EBLearn – C++ Library forEnergy-Based Learning. It includes several demos and step-by-step instructionsto train classifiers based on convolutional neural networks.

·    Torch7 – Provides amatlab-like environment for state-of-the-art machine learning algorithms,including a fast implementation of convolutional neural networks.

·    Deep Learning - Various links for deep learning software.

 

Part-Based Models

·    DeformablePart-based Detector – Libraryprovided by the authors of the original paper (state-of-the-art in PASCAL VOCdetection task)

·    EfficientDeformable Part-Based Detector – Branch-and-Bound implementation for a deformable part-baseddetector.

·    AcceleratedDeformable Part Model – Efficientimplementation of a method that achieves the exact same performance ofdeformable part-based detectors but with significant acceleration (ECCV 2012).

·    Coarse-to-FineDeformable Part Model – Fastapproach for deformable object detection (CVPR 2011).

·    Poselets – C++ and Matlab versions for object detectionbased on poselets.

·    Part-based Face Detector and Pose Estimation – Implementation of a unified approach for facedetection, pose estimation, and landmark localization (CVPR 2012).

 

Attributes and Semantic Features

·         RelativeAttributes – Modifiedimplementation of RankSVM to train Relative Attributes (ICCV 2011).

·         Object Bank – Implementation of object bank semantic features(NIPS 2010). See also ActionBank

·         Classemes, Picodes, and Meta-class features – Software for extracting high-level imagedescriptors (ECCV 2010, NIPS 2011, CVPR 2012).

 

Large-Scale Learning

·         Additive Kernels – Source code for fast additive kernel SVMclassifiers (PAMI 2013).

·         LIBLINEAR – Library for large-scale linear SVMclassification.

·         VLFeat – Implementationfor Pegasos SVM and Homogeneous Kernel map.

 

Fast Indexing and Image Retrieval

·         FLANN – Library for performing fast approximate nearestneighbor.

·         Kernelized LSH – Source code for Kernelized Locality-SensitiveHashing (ICCV 2009).

·         ITQ Binary codes – Code for generation of small binary codes usingIterative Quantization and other baselines such as Locality-Sensitive-Hashing(CVPR 2011).

·         INRIA ImageRetrieval – Efficient code forstate-of-the-art large-scale image retrieval (CVPR 2011).

 

Object Detection

·         See Part-based Models and Convolutional Nets above.

·         PedestrianDetection at 100fps – Very fastand accurate pedestrian detector (CVPR 2012).

·         Caltech Pedestrian Detection Benchmark – Excellent resource for pedestrian detection, with various linksfor state-of-the-art implementations.

·         OpenCV – Enhancedimplementation of Viola&Jones real-time object detector, with trainedmodels for face detection.

·         EfficientSubwindow Search – Sourcecode for branch-and-bound optimization for efficient object localization (CVPR2008).

 

3D Recognition

·         Point-Cloud Library – Libraryfor 3D image and point cloud processing.

 

Action Recognition

·         ActionBank – Source code for action recognition based on theActionBank representation (CVPR 2012).

·         STIP Features – software for computing space-time interest pointdescriptors

·         Independent Subspace Analysis – Look for Stacked ISA for Videos (CVPR 2011)

·         VelocityHistories of Tracked Keypoints - C++ code for activity recognition using the velocity histories oftracked keypoints (ICCV 2009)

 

---

Datasets

Attributes

·         Animals withAttributes – 30,475 images of50 animals classes with 6 pre-extracted feature representations for each image.

·         aYahoo andaPascal – Attributeannotations for images collected from Yahoo and Pascal VOC 2008.

·         FaceTracer – 15,000 faces annotated with 10 attributes andfiducial points.

·         PubFig – 58,797 face images of 200 people with 73attribute classifier outputs.

·         http://vis-www.cs.umass.edu/lfw/ – 13,233 face images of 5,749 people with 73attribute classifier outputs.

·         Human Attributes – 8,000 people with annotated attributes. Check alsothis link for another datasetof human attributes.

·         SUN AttributeDatabase – Large-scale sceneattribute database with a taxonomy of 102 attributes.

·         ImageNetAttributes – Variety ofattribute labels for the ImageNet dataset.

·         Relativeattributes – Data for OSR anda subset of PubFig datasets. Check also this link for the WhittleSearch data.

·         AttributeDiscovery Dataset – Images ofshopping categories associated with textual descriptions.

 

Fine-grained Visual Categorization

·         Caltech-UCSD BirdsDataset – Hundreds of birdcategories with annotated parts and attributes.

·         Stanford DogsDataset – 20,000 images of120 breeds of dogs from around the world.

·         Oxford-IIIT PetDataset – 37 category petdataset with roughly 200 images for each class. Pixel level trimap segmentationis included.

·         Leeds Butterfly Dataset – 832 images of 10 species of butterflies.

·         Oxford FlowerDataset – Hundreds offlower categories.

 

Face Detection

·         http://vis-www.cs.umass.edu/fddb/ – UMass face detection dataset and benchmark(5,000+ faces)

·         CMU/MIT – Classical face detectiondataset.

 

Face Recognition

·         Face Recognition Homepage – Large collection of face recognition datasets.

·         vis-www.cs.umass.edu/lfw – UMassunconstrained face recognition dataset (13,000+ face images).

·         NIST FaceHomepage – includes facerecognition grand challenge (FRGC), vendor tests (FRVT) and others.

·         CMU Multi-PIE – containsmore than 750,000 images of 337 people, with 15 different views and 19 lightingconditions.

·         FERET – Classical face recognition dataset.

·         Deng Cai’s facedataset in Matlab Format – Easy touse if you want play with simple face datasets including Yale, ORL, PIE, andExtended Yale B.

·         SCFace – Low-resolutionface dataset captured from surveillance cameras.

 

Handwritten Digits

·         MNIST – large datasetcontaining a training set of 60,000 examples, and a test set of 10,000examples.

 

Pedestrian Detection

·         Caltech Pedestrian Detection Benchmark – 10 hours of video taken from a vehicle,350K bounding boxes forabout 2.3K unique pedestrians.

·         INRIA PersonDataset – Currently one ofthe most popular pedestrian detection datasets.

·         ETH PedestrianDataset – Urban datasetcaptured from a stereo rig mounted on a stroller.

·         TUD-BrusselsPedestrian Dataset – Datasetwith image pairs recorded in an crowded urban setting with an onboard camera.

·         PASCAL HumanDetection – One of 20categories in PASCAL VOC detection challenges.

·         USC Pedestrian Dataset – Small dataset captured from surveillance cameras.

 

Generic Object Recognition

·         ImageNet – Currentlythe largest visual recognition dataset in terms of number of categories andimages.

·         Tiny Images – 80 million 32x32 low resolution images.

·         Pascal VOC – One of the most influential visual recognitiondatasets.

·         Caltech 101 / Caltech 256 – Popular image datasets containing 101 and 256object categories, respectively.

·         MIT LabelMe – Online annotation tool for building computervision databases.

 

Scene Recognition

·         MIT SUN Dataset – MIT scene understanding dataset.

·         UIUC FifteenScene Categories – Dataset of15 natural scene categories.

 

Feature Detection and Description

·         VGG AffineDataset – Widely useddataset for measuring performance of feature detection and description. CheckVLBenchmarksfor an evaluation framework.

 

Action Recognition

·         Benchmarking Activity Recognition – CVPR 2012 tutorial covering various datasets for actionrecognition.

 

RGBD Recognition

·         RGB-D ObjectDataset – Datasetcontaining 300 common household objects

 

Reference:

[1]: http://rogerioferis.com/VisualRecognitionAndSearch/Resources.html

特征提取

·         SURF特征: http://www.vision.ee.ethz.ch/software/index.de.html(当然这只是其中之一)

·         LBP特征(一种纹理特征)：http://www.comp.hkbu.edu.hk/~icpr06/tutorials/Pietikainen.html

·         Fast Corner Detection（OpenCV中的Fast算法）:FAST CornerDetection -- Edward Rosten

 

机器视觉

·         A simple object detectorwith boosting(Awarded the Best Short Course Prize at ICCV 2005，So了解adaboost的推荐之作)：http://people.csail.mit.edu/torralba/shortCourseRLOC/boosting/boosting.html

·         Boosting(该网页上有相当全的Boosting的文章和几个Boosting代码，本人推荐)：http://cbio.mskcc.org/~aarvey/boosting_papers.html

·         Adaboost Matlab工具：http://graphics.cs.msu.ru/en/science/research/machinelearning/adaboosttoolbox

·         MultiBoost(多类Adaboost算法的程序)：http://sourceforge.net/projects/multiboost/

·         TextonBoost: Jamie Shotton -Code

·         LibSvm（推荐）: http://www.csie.ntu.edu.tw/~cjlin/

·         ConditionalRandom Fields（CRF论文+Code列表，推荐）

·         CRF++: Yet Another CRF toolkit

·         Conditional Random Field (CRF) Toolbox for Matlab

·         Tree CRFs

·         LingPipe:Installation

·         Hidden MarkovModels（推荐）

·         隐马尔科夫模型(Hidden MarkovModels)系列之一 - eaglex的专栏 -博客频道 - CSDN.NET（推荐）

 

综合代码

·         CvPapers(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)：http://www.cvpapers.com/

·         Computer Vision Software(里面代码很多，并详细的给出了分类)：http://peipa.essex.ac.uk/info/software.html

·         某人的Windows Live：https://skydrive.live.com/?cid=3b6244088fd5a769#cid=3B6244088FD5A769&id=3B6244088FD5A769!523

·         MATLAB and OctaveFunctions for Computer Vision and Image Processing（这个里面的东西也很全，只是都是用Matlab和Octave开发的）：http://www.csse.uwa.edu.au/~pk/research/matlabfns/

·         Computer Vision Resources（里面的视觉算法很多，给出了相应的论文和Code，挺好的）：https://netfiles.uiuc.edu/jbhuang1/www/resources/vision/index.html

·         MATLAB Functions forMultiple View Geometry（关于物体多视角计算的库）：http://www.robots.ox.ac.uk/~vgg/hzbook/code/

·         Evolutive Algorithm basedon Naïve Bayes models Estimation（单独列了一个算法的Code）：http://www.cvc.uab.cat/~xbaro/eanbe/#_Software

 

主页代码

·         Pablo Negri'sHome Page

·         Jianxin Wu'shomepage

·         Peter Carbonetto

·         Markov Random Fields for Super-Resolution

·         Detecting andSketching the Common

·         PedroFelzenszwalb

·         Hae JONG, SEO

·         CAP 5416 -Computer Vision

·         ParallelTracking and Mapping for Small AR Workspaces (PTAM)

·         Deva Ramanan - UC Irvine - Computer Vision

·         RaghuramanGopalan

·         Hui Kong

·         Jamie Shotton -Post-Doctoral Researcher in Computer Vision

·         Jean-YvesAUDIBERT

·         Olga Veksler

·         Stephen Gould

·         Publications(Last Update: 09/30/10)

·         Karim Ali -FlowBoost

·         A simple parts and structure object detector

·         Code - OxfordBrookes Vision Group

·         Taku Kudo

 

行人检测

·         Histogram ofOriented Gradient (Windows)

·         INRIA Pedestriandetector

·         Poselets

·         William RobsonSchwartz - Softwares

·         calvin upper-body detector v1.02

·         RPT@CVG

·         Main Page

·         Source Code

·         Dr. Luciano Spinello

·         PedestrianDetection

·         Class-Specific Hough Forests for Object Detection

·         Jianxin Wu'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：http://www.cs.berkeley.edu/~smaji/projects/ped-detector/

 

视觉壁障

·         High SpeedObstacle Avoidance using Monocular Vision and Reinforcement Learning

·         TLD(2010年很火的tracking算法)

·         online boostingtrackers

·         Boris Babenko

·         Optical Flow AlgorithmEvaluation (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)http://of-eval.sourceforge.net/

 

物体检测算法

·         Object Detection

·         Software for object detection

 

人脸检测

·         Source Code

·         10个人脸检测项目

·         Jianxin Wu'shomepage

 

ICA独立成分分析

·         An ICA page-papers,code,demo,links (Tony Bell)

·         FastICA

·         Cached k-d tree search for ICP algorithms

 

滤波算法

·         卡尔曼滤波：The KalmanFilter(推荐)

·         Bayesian FilteringLibrary: The BayesianFiltering Library

 

路面识别

·         Source Code

·         Vanishing pointdetection for general road detection

 

分割算法

·         MATLAB Normalized CutsSegmentation Code：software

·         超像素分割：SLIC Superpixels

 

ZZ： http://blog.sina.com.cn/s/blog_5086c3e20101kdy5.html，http://www.yuanyong.org/cv/cv-cc