本帖最后由 mindvision 于 2015-11-18 16:57 编辑
机器视觉开源代码集合 · Geometric Blur [7] [ Code] · Global and EfficientSelf-Similarity [9] [ Code] · Pyramids of Histogramsof Oriented Gradients [ Code] · Space-Time InterestPoints (STIP) [14][ Project] [ Code] · Boundary PreservingDense Local 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] · 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] · Online Selection ofDiscriminative Tracking Features[ Project]
二、图像分割Image Segmentation:
· OWT-UCM HierarchicalSegmentation [5] [ Resources] · 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] · Ef?cient 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 ExperimentalComparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in ComputerVision[ Paper][ Code] · Geodesic Star Convexityfor Interactive Image Segmentation[ Project] · Contour Detection andImage Segmentation Resources[ Project][ Code] · 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 MethodResearch by Chunming Li[ Project] · ClassCut forUnsupervised Class 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] · 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] · 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 viaInformation Maximization [4] [ Matlab code] · Saliency detection: Aspectral residual approach. [7] [ Matlab code] · Segmenting salientobjects from images and videos. [8] [ Matlab code] · Discriminant Saliencyfor Visual Recognition from Cluttered Scenes. [10] [ Code] · Learning to PredictWhere Humans Look [11] [ Project] · Global Contrast basedSalient Region Detection [12] [ Project] · Bayesian Saliency viaLow and 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
· Spatial Pyramid Matching[2] [ Code] · Texture Classification[5] [ Project] · Multiple Kernels forImage Classification [6] [ Project] · Large Scale CorrelationClustering Optimization[ Matlab code] · Detecting and Sketchingthe Common[ Project] · User Assisted Separationof Reflections from a Single Image Using a Sparsity Prior[ Paper][ Code] · Filters for TextureClassification[ Project] · Multiple Kernel Learningfor Image Classification[ Project]
六、抠图Image Matting
· A Closed Form Solutionto Natural Image Matting [ Code] · 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] · 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] · the HandVu:vision-based hand gesture interface[ Project] · Learning ProbabilisticNon-Linear Latent Variable Models for Tracking Complex Activities[ Project]
八、Kinect:
· 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] · 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:
· Probabilistic LatentSemantic Analysis (pLSA)[ Code] · FASTANN and FASTCLUSTERfor approximate k-means (AKM)[ Project] · Fast Intersection /Additive Kernel SVMs[ Project] · Deep Learning Methodsfor Vision[ Project] · Neural Network forRecognition of Handwritten Digits[ Project] · Training a deepautoencoder or a classifier on MNIST digits[ Project] · THE MNIST DATABASE ofhandwritten digits[ Project] · Ersatz:deep neural networks in the cloud[ 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] · 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's HomePage[ 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, ActionRecognition:
· Action Recognition byDense Trajectories[ Project][ Code] · Action Recognition Usinga Distributed Representation of Pose and Appearance[ Project] · 2D Articulated HumanPose Estimation[ Project] · Fast Human PoseEstimation Using Appearance and Motion via Multi-Dimensional BoostingRegression[ Paper][ Code] · Estimating Human Posefrom Occluded Images[ Paper][ Code] · Quasi-dense widebaseline matching[ 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 humanpose estimation 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、人手及指尖检测与识别:
· finger-detection-and-gesture-recognition[ Code] · Hand and FingerDetection using 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 Implementationof the 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] · Beyond Pixels: ExploringNew Representations and Applications for Motion Analysis[ Project] · A Database and EvaluationMethodology for Optical Flow[ Project] · Robust Optical FlowEstimation [ 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 ObjectExtraction, 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 DescriptionGeneral Libraries:
· VLFeat –Implementation of various feature descriptors (including SIFT, HOG, and LBP)and covariant feature detectors (including DoG, Hessian, Harris Laplace,Hessian Laplace, Multiscale Hessian, Multiscale Harris). Easy-to-use Matlabinterface. See Modern features: Software –Slides providing a demonstration of VLFeat and also links to other software.Check also VLFeat hands-on session training· OpenCV –Various implementations of modern feature detectors and descriptors (SIFT,SURF, FAST, BRIEF, ORB, FREAK, etc.)
Fast Keypoint Detectors for Real-time Applications:
· FAST – High-speed corner detector implementation for a wide variety ofplatforms · AGAST – Even faster than the FAST cornerdetector. A multi-scale version of this method is used for the BRISK descriptor(ECCV 2010).
Binary Descriptors for Real-Time Applications:
· BRIEF – C++ code for a fast and accurate interestpoint descriptor (not invariant to rotations and scale) (ECCV 2010) · ORB –OpenCV implementation of the Oriented-Brief (ORB) descriptor (invariant torotations, 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:
Other Local Feature Detectors and Descriptors:
· LIOP descriptor –Source code for the Local Intensity order Pattern (LIOP) descriptor (ICCV2011). · Local Symmetry Features – Source code formatching of local symmetry features under large variations in lighting, age,and rendering style (CVPR 2012).
Global Image Descriptors:
· GIST –Matlab code for the GIST descriptor · CENTRIST – Global visual descriptor for scenecategorization and object detection (PAMI 2011)
Feature Coding and Pooling
· VGG Feature Encoding Toolkit – Source code forvarious state-of-the-art feature encoding methods – including Standard hardencoding, Kernel codebook encoding, Locality-constrained linear encoding, andFisher kernel encoding. · Spatial Pyramid Matching – Source code forfeature pooling based on spatial pyramid matching (widely used for imageclassification)
Convolutional Nets and Deep Learning
· EBLearn –C++ Library for Energy-Based Learning. It includes several demos andstep-by-step instructions to train classifiers based on convolutional neuralnetworks. · Torch7 –Provides a matlab-like environment for state-of-the-art machine learningalgorithms, including a fast implementation of convolutional neural networks.
Part-Based Models
· Accelerated Deformable Part Model – Efficientimplementation of a method that achieves the exact same performance ofdeformable part-based detectors but with significant acceleration (ECCV 2012). · Poselets – C++ and Matlab versions for objectdetection based on poselets.
Attributes and Semantic Features
· Relative Attributes – Modified implementation of RankSVM totrain Relative Attributes (ICCV 2011).
Large-Scale Learning
· Additive Kernels – Source code for fast additive kernel SVMclassifiers (PAMI 2013). · LIBLINEAR – Library for large-scale linear SVMclassification. · VLFeat –Implementation for Pegasos SVM and Homogeneous Kernel map.
Fast Indexing and Image Retrieval
· FLANN – Library for performing fast approximatenearest neighbor. · Kernelized LSH – Source code for KernelizedLocality-Sensitive Hashing (ICCV 2009). · ITQ Binary codes – Code for generation of small binary codesusing Iterative Quantization and other baselines such asLocality-Sensitive-Hashing (CVPR 2011).
Object Detection
· OpenCV –Enhanced implementation of Viola&Jones real-time object detector, withtrained models for face detection.
3D Recognition
Action Recognition
· ActionBank – Source code for action recognition basedon the ActionBank representation (CVPR 2012). · STIP Features – software for computing space-timeinterest point descriptors
DatasetsAttributes
· Animals with Attributes – 30,475 images of 50animals classes with 6 pre-extracted feature representations for each image. · aYahoo and aPascal – Attribute annotations for imagescollected from Yahoo and Pascal VOC 2008. · FaceTracer – 15,000 faces annotated with 10 attributesand fiducial points. · PubFig – 58,797 face images of 200 people with 73attribute classifier outputs. · [url=http://vis- www.cs.umass.edu/lfw/]LFW[/url] –13,233 face images of 5,749 people with 73 attribute classifier outputs. · Human Attributes – 8,000 people with annotated attributes.Check also this link for another dataset of human attributes. · Relative attributes – Data for OSR and a subset of PubFig datasets.Check also this link for the WhittleSearch data.
Fine-grained Visual Categorization
· Oxford-IIIT Pet Dataset – 37 category petdataset with roughly 200 images for each class. Pixel level trimap segmentationis included.
Face Detection
· [url=http://vis- www.cs.umass.edu/fddb/]FDDB[/url] – UMass face detection dataset and benchmark (5,000+ faces) · CMU/MIT –Classical face detection dataset.
Face Recognition
· [url=http://vis- www.cs.umass.edu/lfw/]LFW[/url] –UMass unconstrained face recognition dataset (13,000+ face images). · NIST Face Homepage – includes face recognition grand challenge(FRGC), vendor tests (FRVT) and others. · CMU Multi-PIE –contains more than 750,000 images of 337 people, with 15 different views and 19lighting conditions. · FERET – Classical face recognition dataset. · SCFace –Low-resolution face dataset captured from surveillance cameras.
Handwritten Digits
· MNIST – large dataset containing a training setof 60,000 examples, and a test set of 10,000 examples.
Pedestrian Detection
Generic Object Recognition
· ImageNet –Currently the largest visual recognition dataset in terms of number ofcategories and images. · Tiny Images – 80 million 32x32 low resolution images. · Pascal VOC – One of the most influential visualrecognition datasets. · MIT LabelMe – Online annotation tool for buildingcomputer vision databases.
Scene Recognition
Feature Detection and Description
· VGG Affine Dataset – Widely used dataset for measuringperformance of feature detection and description. Check VLBenchmarksfor an evaluation framework.
Action Recognition
RGBD Recognition
欢迎来电咨询或访问迈德威视官网 http://www.mindvision.com.cn
迈德威视的相机无缝兼容halcon、opencv、labview等机器视觉软件,为您提供最具性价比的工业相机。
电话:0755-28376557 手机: 18576735596 李工 QQ: 2390812618 |