Computer Vision Papers By Topic

Cynbe ru Taren

For a motivational overview see What Can We Learn From Computer Vision?

For a human angle, see some of my favorite computer vision researchers.

Here are a few hundred of the most significant computer vision papers, by topic. (Also available as a simple chronological list.)

NB: You may find Google Scholar useful when tracking down references; other useful resources include CiteSeer and arXiv.

Topics:

• Ancillary Algorithms and Datastructures
• Belief Propagation
• Boosting
• Compressive Sampling
• Cows
• Dynamic Programming
• Face Expressions
• Face Detection
• Face Recognition
• Geometry Representation
• Graph Cut
• Graph Algorithms
• Hashing Techniques
• Histogram Techniques
• Hidden Markov Models
• Hough Transform
• Image Analysis
• Image Differencing
• Image Matching and Restoration, Optical Flow and Stereo Vision
• Image Segmentation and Background Models
• Isoperimetric Algorithms
• Iterated Conditional Modes
• Level Sets
• Linear Programming
• Mathematical Background
• Mean Shift, Adaptive Mean Shift, Continuously Adapative Mean Shift ("CamShift")
• Motion, Tracking and Track Analysis
• New View Synthesis
• Nonlinear Optimization
• Object Detection and Recognition, and Post Estimation
• Occlusion Handling
• Oliphants
• Particle Filtering
• Random Fields
• Robust Regresssion: RANSAC and Kith
• Semi-Definite Programming
• Semi-Global Matching
• Shadow Handling
• Smoothing
• Super Resolution
• Swendsen-Wang
• Textures
• Thresholding
• Video Editing

• Ancillary Algorithms and Datastructures:
  • Brub08: Isotropic PCA and Affine-Invariant Clustering
  • Chen08: Linear Time Recognition Algorithms for Topological Invariants in 3D
  • Mair08: Supervised Dictionary Learning
  • Mant08: Improved Smoothed Analysis of the k-Means Method: "by far the most popular..."
  • Mohi08: A fast approach for overcomplete sparse decomposition based on smoothed L0 norm
  • Rost08: Camera distortion self-calibration using the plumb-line constraint and minimal Hough entropy
  • Vasi08: Learning Isometric Separation Maps
  • Orli07: A faster strongly polynomial time algorithm for submodular function minimization
  • Schl06: Transforming an arbitrary minsum problem into a binary one
  • Felz04c: Distance Transforms of Sampled Functions
  • Smin04: Generative modeling for continuous non-linearly embedded visual inference
  • Scho00: Video textures
  • Schr00: A combinatorial algorithm minimizing submodular functions in strongly polynomial time
  • Chia92: Dynamic algorithms in computational geometry
  • Cohe91: Dynamic expression trees and their applications
  • Fred87: Fibonacci heaps and their uses in improved network optimization algorithms

• Belief Propagation and Tree-Reweighted Message Passing (compare to Graph cuts, Isoperimetric, Nonlinear optimization, Semi-Global Matching and Swendsen-Wang). In general, belief propagation is slower but more powerful than graph cuts, so it tends to be used on smaller, harder problems, often in combination with graph cuts for lower-level processing:
  • Alah08: Reduce, Reuse & Recycle: Efficiently Solving Multilabel MRFs.
  • Bick08: Distributed Kalman Filter via Gaussian Belief Propagation
  • Bick08b: Polynomial Linear Programming with Gaussian Belief Propagation
  • Bick08c: A Gaussian Belief Propagation Solver for Large Scale Support Vector Machines
  • Bick08d: Gaussian Belief Propagation Based Multiuser Detection
  • Kuma08: Learning Layered Motion Segmentations of Video.
  • Kuma08b: Efficiently Solving Convex Relaxations for MAP Estimation: Improving on TRW; guaranteed convergence on non-submodular fns
  • Sang08: Message-passing for Maximum Weight Independent Set
  • Shen08: Gaussian Belief Propagation Solver for Systems of Linear Equations
  • Sont08: Tightening LP Relaxations for MAP using message passing
  • Bhat07: Using Photographs to Enhance Videos of a Static Scene
  • Weis07b: MAP Estimation, Linear Programming and Belief Propagation with Convex Free Energies: Provable MAP extraction.
  • Yang06: Stereo Matching with Color-Weighted Correlation, Hierachical Belief Propagation and Occlusion Handling
  • Kolm06: Comparison of energy minimization algorithms for highly connected graphs
  • Kuma06: Solving Markov Random Fields using Second Order Cone Programming Relaxations
  • Levi06: Learning to combine bottom-up and top-down segmentation
  • Szel06: A Comparative Study of Energy Minimization Methods for Markov Random Fields
  • Boim05: Detecting Irregularities in Images and in Video
  • Kolm05b: Convergent Tree-Reweighted Message Passing for Energy Minimization
  • Lan05b: Efficient Belief Propagation with Learned Higher-Order Markov Random Fields
  • Melt05: Globally optimal solutions for energy minimization in stereo vision using reweighted belief propagation
  • Siga04: Tracking loose-limbed people
  • Kolm05b: Convergent Tree-Reweighted Message Passing for Energy Minimization: Improves on Wain03 via better convergence properties.
  • Wain05: MAP Estimation Via Agreement on Trees: Message-Passing and Linear Programming
  • Felz04: Efficient Belief Propagation for Early Vision: Radical speed-ups.
  • Kuma04: Extending pictorial structures for object recognition
  • Yano04: Finding the M Most Probable Configurations Using Loopy Belief Propagation
  • Levi03: Learning How to Inpaint from Global Image Statistics
  • Rama03: Finding and Tracking People from the Bottom Up
  • Siga03: Attractive people: Assembling loose-limbed models using non-parametric belief propagation
  • Tapp03: Comparison of Graph Cuts with Belief Propagations for Stereo, using Identical MRF Parameters
  • Wain03: Tree-reweighted belief propagation algorithms and approximate ML estimation by pseudo-moment matching: The original paper.
  • Sudd02: Nonparametric Belief Propagation and Facial Appearance Estimation
  • Sun02: Stereo Matching Using Belief Propagation
  • Yedi01: Constructive Free Energy Approximations and Generalized Belief Propagaion Algorithms
  • Viol01: Rapid Object Detection using a Boosted Cascade of Simple Features: "first highly accurate as well as real-time frontal face detector"
  • Yedi00: Constructive Free Energy Approximations and Generalized Belief Propagation Algorithms
  • Free00: Learning Low-Level Vision : Seminal loopy belief propagation paper..
  • Smyt98: Belief Networks, Hidden Markov Models, and Markov Random Fields: a Unifying View

• Boosting:
  • Shen09: A Duality View of Boosting Algorithms
  • Hou07: Multiview Pedestrian Detection Based on Vector Boosting
  • Nowo07: Weighted Substructure Mining for Image Analysis
  • Tuze07: Human Detection via Classification on Riemannian Manifolds
  • Yin07: Tree-based Classifiers for Bilayer Video Segmentation
  • Grab06: Real-Time Tracking via On-line Boosting
  • Grab06b: Online Boosting and Vision
  • Lapt06: Improvements of Object Detection Using Boosted Histograms
  • Le06: Ent-Boost: Boosting using entropy measures for robust object detection
  • Huan05: Vector Boosting for Rotation Invariant Multi-View Face Detection
  • Oze05: Online Bagging and Boosting
  • Serr05: Object Recognition with Features Inspired by Visual Cortex
  • Wolf05: Robust Boosting for Learning from Few Examples
  • Wu05: Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors
  • Cris04: A Multi-Stage Approach to Facial Feature Detection
  • Huan04: Boosting Nested Cascade Detector for Multi-View Face Detection
  • Huan04b: Omni-Directional Face Detection Based on Real AdaBoost
  • Levi04: Learning Object Detection from a Small Number of Examples: the Importance of Good Features
  • Torr04: Sharing Visual Features for Multiclass and Multiview Object Detection
  • Torr04b: Sharing features: efficient boosting procedures for multiclass object detection
  • Lien03: A Detector Tree of Boosted Classifiers for Real-time Object Detection and Tracking
  • Scha03: The Boosting Approach to Machine Learning: Book chapter.
  • Viol03: Detecting Pedestrians Using Patterns of Motion and Appearance
  • Maso00: Boosting Algorithms as Gradient Descent
  • Cohe99: A Simple, Fast and Effective Rule Learner Freu99: A Short Introduction to Boosting.
  • Frie98: Additive Logistic Regression: a Statistical View of Boosting: Original gentleBoost paper)
  • Scha98: Improved Boosting Algorithms Using Confidence-rated Predictions: Original Real AdaBoost paper,
  • Freu95: A decision-theoretic generalization of on-line learning and an application to boostingOriginal Discrete AdaBoost ("adaptive boost") paper,

• Compressive Sampling (see also Compressive Sensing Resources):
  • Bara07: Compressive sensing.: tutorial
  • Yang07: Feature Selection in Face Recognition: A Sparse Representation Perspective
  • Cand06: Compressive sampling tutorial

• Cows:
  • Levi06: Learning to combine bottom-up and top-down segmentation
  • Schr06: Single-Histogram Class Models for Image Segmentation
  • Shot06: Textonboost: Joint appearance, shape and context modeling for multi-class object recognition and segmentation
  • Kuma05: OBJCUT: pictorial structures, combining top-down + bottom-up cues
  • Kuma04: Extending pictorial structures for object recognition

• Oliphants:
  • Wang05: Interactive video cutout

• Dynamic Programming:
  • Kolm05e: Bi-layer segmentation of binocular stereo video
  • Felz00: Efficient Matching of Pictorial Structures

• Face expressions:
  • Levy08: Data-Driven Enhancement of Facial Attractiveness
  • Perl06: Face Recognition using Principal Component Analysis and Log-Gabor Filters
  • Perl06b: Recognition of expression variant faces using masked log-Gabor features and Principal Component Analysis
  • Tu05: Image Parsing: Unifying Segmentation, Detection, and Recognition
  • Pari04: Extraction of Three-dimensional Information from Images: dissertation; face relighting; hair geometry capture.
  • Sudd02: Nonparametric Belief Propagation and Facial Appearance Estimation
  • Pigh98: Synthesizing realistic facial expressions from photographs

• Face detection:
  • Mogh08: Sparse Regression as a Sparse Eignvalue Problem
  • Mogh07: Fast Pixel/Part Selection with Sparse Eigenvectors: 1000X faster; eigenfaces; gender classification.
  • Osad07: Synergistic Face Detection and Pose Estimation with Energy-Based Models
  • Riha06: OBJCUT for face detection
  • Berg05: Shape matching and object recognition using low distortion correspondence
  • Hamp05: Smart Video Surveillance (magazine survey article)
  • Huan05: Vector Boosting for Rotation Invariant Multi-View Face Detection
  • Tu05: Image Parsing: Unifying Segmentation, Detection, and Recognition
  • Cris04: A Multi-Stage Approach to Facial Feature Detection
  • Huan04: Boosting Nested Cascade Detector for Multi-View Face Detection
  • Huan04b: Omni-Directional Face Detection Based on Real AdaBoost
  • Opel04: Generic Object Recognition with Boosting
  • Pati04: People Detection and Tracking in High Resolution Panoramic Video Mosaic
  • Viol04: Robust Real-Time Face Detection
  • Coma03: Kernel-Based Object Tracking: adaptive mean shift.
  • Jone03: Fast Multi-view Face Detection
  • Brad02: Computer Vision Face Tracking For Use in a Perceptual User Interface
  • Leun01: Representing and Recognizing the Visual Appearance of Materials using Three-dimensional Textons
  • Viol01: Rapid Object Detection using a Boosted Cascade of Simple Features: "first highly accurate as well as real-time frontal face detector"
  • Schn00: A Statistical Method for 3D Object Detection Applied to Faces and Cars
  • Stor99: Skin colour detection under changing lighting conditions

• Face recognition:
  • Blai08: Culture Shapes How We Look at Faces
  • Koki08: Classification of multiple observations by semi-supervised learning
  • Llon08: 3D Face Recognition with Sparse Spherical Representations
  • Bolm07: FacePerf: Benchmarks for Face Recognition Algorithms with source code
  • Kuma07b: An Invariant Large Margin Nearest Neighbour Classifier: rotation invariance, face recognition, convex SDP.
  • Otoo07: Face Recognition Algorithms Surpass Humans Matching Faces over Changes in Illumination
  • Yang07: Feature Selection in Face Recognition: A Sparse Representation Perspective
  • Ahon06: Face Description with Local Binary Patterns: Application to Face Recognition
  • Liu06: Capitalize on Dimensionality Increasing Techniques for Improving Face Recognition Grand Challenge Performance
  • Perl06: Face Recognition using Principal Component Analysis and Log-Gabor Filters
  • Perl06b: Recognition of expression variant faces using masked log-Gabor features and Principal Component Analysis
  • Phil06: Preliminary Face Recognition Grand Challenge Results
  • Sinh06: Face Recognition by Humans: Nineteen Results All Computer Vision Researchers Should Know About
  • Aran05: Automatic Face Recognition for Film Character Retrieval in Feature-Length Films
  • Aran05b: Face recognition with image sets using manifold density divergence
  • Berg05: Shape matching and object recognition using low distortion correspondence
  • Bowy05: A survey of approaches and challenges in 3D and multi-modal 3D + 2D face recognition
  • Bron05: Three-Dimensional Face Recognition
  • He05: Face Recognition Using Laplacianfaces
  • Phil05: Overview of the Face Recognition Grand Challenge
  • Torr05: Representational oriented component analysis (ROCA) for face recognition with one sample image per Training Class
  • Ahon04: Face recognition with local binary patterns
  • Ho03: Clustering appearances of objects under varying illumination conditions
  • Zhao03: Face Recognition A Literature Survey
  • Nevi99: Face Recognition Using an Embedded HMM

• Feature Detectors and Wavelets:
  • Cour08: Fast Density Codes for Image Data
  • Weis07: What makes a good model of natural images?
  • Ahon06: Face Description with Local Binary Patterns: Application to Face Recognition
  • Ahon04: Face recognition with local binary patterns
  • Miko06: Multiple Object Class Detection with a Generative Model
  • Nowa06: Sampling Strategies for Bag-of-Features Image Classification
  • Perl06: Face Recognition using Principal Component Analysis and Log-Gabor Filters
  • Perl06b: Recognition of expression variant faces using masked log-Gabor features and Principal Component Analysis
  • Russ06: Using multiple segmentations to discover objects and their extent in image collections
  • Shot06: Textonboost: Joint appearance, shape and context modeling for multi-class object recognition and segmentation
  • Skre06: Texture classification using sparse frame-based representations
  • Zhan06: Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study
  • Miko05: A Performance Evaluation of Local Descriptors: SIFT wins; improved SIFT.
  • Roth05b: A framework for learning image priors
  • Serr05: Object Recognition with Features Inspired by Visual Cortex
  • Varm05: A Statistical Approach to Texture Classification from Single Images
  • Wu05: Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors
  • Huan04: Boosting Nested Cascade Detector for Multi-View Face Detection
  • Huan04b: Omni-Directional Face Detection Based on Real AdaBoost
  • Levi04: Learning Object Detection from a Small Number of Examples: the Importance of Good Features
  • Opel04: Generic Object Recognition with Boosting
  • Sten04b: Hand pose estimation using hierarchical detection
  • Torr04: Sharing Visual Features for Multiclass and Multiview Object Detection
  • Torr04b: Sharing features: efficient boosting procedures for multiclass object detection
  • Sten04: Model-Based hand Tracking Using a Hierarchical Basyesian Filter: thesis
  • Coma03: Kernel-Based Object Tracking: adaptive mean shift.
  • Ferg03: Object Class Recognition by Unsupervised Scale-Invariant Learning
  • Jone03: Fast Multi-view Face Detection
  • Lien03: A Detector Tree of Boosted Classifiers for Real-time Object Detection and Tracking
  • Lowe03: Distinctive Image Features from Scale-Invariant Keypoints
  • Varm03: Texture Classification: Are Filter Banks Necessary?
  • Viol03: Detecting Pedestrians Using Patterns of Motion and Appearance
  • Gree02: Canny Edge Detection Tutorial
  • Grig02: Comparison of Texture Features Based on Gabor Filters
  • Lien02: An extended set of Haar-like features for rapid object detection
  • Thay02: Shape context and chamfer matching in cluttered scenes
  • Zhu02: What are Textons?
  • Berg01: Geometric blur for template matching
  • Leun01: Representing and Recognizing the Visual Appearance of Materials using Three-dimensional Textons
  • Schn00: A Statistical Method for 3D Object Detection Applied to Faces and Cars
  • Mihc99: Low-complexity image denoising based on statistical modeling of wavelet coefficients
  • Stor99: Skin colour detection under changing lighting conditions
  • Torr84: On Edge Detection

• Geometry Representation
  • Pata08: A Nonparametric Approach to 3D Shape Analysis from Digital Camera Images Rava08: Unwrap Mosaics: A new representation for video editing
  • Leib07: Robust Object Detection with Interleaved Categorization and Segmentation
  • Meta07: Physics-Based Modeling Of Nonrigid Objects For Vision and Graphics
  • Grau06: Approximate Correspondences in High Dimensions
  • Felz05: Pictorial Structures for Object Recognition
  • Peyr04: Geodesic Remeshing Using Front Propagation
  • Matu01: Polyhedral Visual Hulls for Real-Time Rendering
  • Amen00: Accurate and Efficient Unions of Balls: fast collision detection.
  • Felz00: Efficient Matching of Pictorial Structures
  • Szel91: Surface Modeling with Oriented Particle Systems

• Graph cuts (compare to Belief Propagation, Isoperimetric, Nonlinear optimization and Swendsen-Wang). Boyk02 gives the currently most popular algorithm. As usual, dissertations provide the easiest way to get oriented, so I suggest starting off by reading Veks99, Kolm03 and Kohl07c. Since much of the older stuff is now of mainly historical interest, your best bet from there is to find a recent paper that interests you, and then read its references as necessary:
  • Alah08: Reduce, Reuse & Recycle: Efficiently Solving Multilabel MRFs.
  • Chan08: A Computational Study of the Pseudoflow and Push-relabel Algorithms for the Maximum Flow Problem.
  • Delo08: A Scalable Graph-Cut Algorithm for N-D Grids
  • Ding08: Two-Frames Accurate Motion Segmentation Using Tensor Voting and Graph Cuts
  • Khol08: Robust Higher Order Potentials for Enforcing Label Consistency.
  • Khol08a: Graph cuts for minimizing robust higher order potentials.
  • Kuma08: Learning Layered Motion Segmentations of Video.
  • Perc08: The Peculiar Phase Structure of Random Graph Bisection
  • Rama08: Exact Inference in Multi-label CRFs with Higher Order Cliques
  • Babe07: Experimental Evaluation of Parametric Max-Flow Algorithms
  • Camp07: Automatic 3D Object Segmentation in Multiple Views using Volumetric Graph-Cuts
  • Gupt07: Efficient inference with cardinality-based clique potentials "orders of magnitude faster than TRW and graph-cut"
  • Hern07: Probabilistic visibility for multi-view stereo
  • Kim07: A new graph cut-based multiple active contour algorithm without initial contours and seed points
  • Kohl07: P3 & Beyond: Solving Energies with Higher Order Cliques
  • Kohl07b: Dynamic Graph Cuts for Efficient Inference in Markov Random Fields
  • Kohl07c: Minimizing Dynamic and Higher Order Energy Functions using Graph Cuts
  • Kolm07: Applications of parametric maxflow in computer vision
  • Lemp07: Global Optimization for Shape Fitting
  • Lemp07b: LogCut-Efficient Graph Cut Optimization for Markov Random Fields: "order of magnitude speedup"
  • Roth07: Optimizing Binary MRFs via Extended Roof Duality
  • Shar07: A Practical Approach for Super-Resolution using Photometric Stereo and Graph Cuts
  • Sinh07: Multi-View Stereo via Graph Cuts on the Dual of an Adaptive Tetrahedral Mesh
  • Thor07: Fully-Dynamic Min-Cut
  • Veks07: Graph Cut Based Optimization for MRFs with Truncated Convex Priors: range move
  • Vogi07: Multi-view Stereo via Volumetric Graph-cuts and Occlusion Robust Photo-Consistency
  • Weis07: What makes a good model of natural images?
  • Yin07: Tree-based Classifiers for Bilayer Video Segmentation
  • Appl06: Globally minimal surfaces by continuous maximal flows
  • Boyk06: An Integral Solution to Surface Evolution PDEs via Geo-Cuts
  • Boyk06b: From Photohulls to Photoflux Optimization
  • Boyk06c: Graph Cuts and Efficient NDImage Segmentation
  • Bray06: PoseCut: Simultaneous Segmentation and 3D Pose Estimation of Humans using Dynamic Graph-Cuts.
  • Crem06: Statistical Priors for Efficient Combinatorial Optimization via Graph Cuts "hundreds of constraints per pixel"
  • Crim06: Bilayer Segmentation of Live Video
  • Grub06: Incorporating non-motion cues into 3D motion segmentation
  • Heng06: Rapid Interactive Modelling from Video with Graph Cuts: dynamic graph cuts
  • Juan06: Active Graph Cuts: Speed-ups via good initial cut estimate.
  • Kohl06: Measuring Uncertainty in Graph Cut Solutions
  • Kolm06: Comparison of energy minimization algorithms for highly connected graphs: min-cut beats TRW + BP.
  • Kuma06: Solving Markov Random Fields using Second Order Cone Programming Relaxations
  • Lemp06: Oriented visibility for multiview reconstruction
  • Levi06: Learning to combine bottom-up and top-down segmentation
  • Pari06: A Surface Reconstruction Method Using Global Graph Cut Optimization
  • Raj06: MRFs for MRIs: Bayesian reconstruction of MR images via graph cuts
  • Riha06: OBJCUT for face detection
  • Roth06: Cosegmentation of Image Pairs by Histogram Matching -- Incorporating a Global Constraint into MRFs
  • Scho06: Near real-time motion segmentation using graph cuts
  • Stic06: Global Depth from Epipolar Volumes--A General Framework for Reconstructing Non-Lambertian Surfaces "reconstruct highly specular surfaces"
    
  • Sun06: Using Strong Shape Priors for Stereo
  • Sun06b: Background cut: background contrast attenuation; min-cut; video teleconferencing.
  • Szel06: A Comparative Study of Energy Minimization Methods for Markov Random Fields
  • Wu06b: Dense Photometric Stereo: A Markov Random Field Approach
  • Deng05: A symmetric patch-based correspondence model for occlusion handling
  • Free05: Energy minimization via graph cuts: Settling what is possible
  • Kolm05c: What Metrics Can Be Approximated by Geo-Cuts, or Global Optimization of Length/Area and Flux
  • Kolm05d: On the optimality of tree-reweighted max-product message passing
  • Kolm05e: Bi-layer segmentation of binocular stereo video
  • Kolm05f: Primal-dual Algorithm for Convex Markov Random Fields: "for the panoramic stitching problem our method outperforms other techniques"
  • Kolm05g: Graph Cut Algorithms for Binocular Stereo with Occlusions
  • Komo05: A new framework for approximate labeling via graph cuts: generalizes alpha-expansion; primal-dual schema.
  • Kuma05: OBJCUT: pictorial structures, combining top-down + bottom-up cues
  • Li05: Video object cut and paste
  • Mats05: Practical Efficiency of Maximum Flow Algorithms Using MA Ordering and Preflows
  • Rico05: Large scale stereo reconstruction by a minimum s-t cut formulationi
  • Raj05: A Graph Cut Algorithm for Generalized Image Deconvolution: MRI reconstruction
  • Roth05: Digital Tapestry: Extending expansion move with non-metric hard and soft constraints.
  • Shei05: Bayesian object detection in dynamic scenes: pixel correlations, temporal persistence, foreground modelling
  • Sinh05: Multi-view Reconstruction using Photo-consistency and Exact Silhouette Constraints: A Maximum-Flow Formulation
  • Vogi05: Multi-view Stereo via Volumetric Graph-cuts
  • Wang05: Interactive video cutout
  • Wern05: Image Segmenation Using Minimum st Cut
  • Xiao05: Motion Layer Extraction in the Presence of Occlusion using Graph Cuts
  • Yuan05: Volume cutout: "novel two-pass graph-cuts algorithm"
  • Zhan05: Simultaneous Parametric Maximum Flow Algorithm with Vertex Balancing
  • Agar04: Interactive digital photomontage
  • Blak04: Interactive image segmentation using an adapative GMMRF model
  • Boyk04: Optimal Object Extraction via Constrained Graph-Cuts (No copy available online)
  • Felz04: Efficient Belief Propagation for Early Vision
  • Hong04: Segment-based stereo matching using graph cuts
  • Howe04: Better Foreground Segmentation Through Graph Cuts
  • Kirs04: A discrete global minimization algorithm for continuous variational problems
  • Mude04: Super resolution using graphcut
  • Pang04: A sentimental education: sentiment analysis using subjectivity summarization based on minimum cuts: Text analysis via min-cut.
  • Pari04: Extraction of Three-dimensional Information from Images: dissertation; face relighting; hair geometry capture.
  • Roth04: "GrabCut": interactive foreground extraction using iterated graph cuts
  • Xiao04: Motion Layer Extraction in the Presence of Occlusion Using Graph Cuts: extracts affine transformations + motion layers.
  • Zabi04: Spatially coherent clustering using graph cuts: "operates simultaneously in feature and image space"
  • Boyk03: Computing geodesics and minimal surfaces via graph cuts
  • Fuji03: New Maximum Flow Algorithms by MA Orderings and Scaling
  • Gold03: Joint 3D-reconstruction and background separation in multiple views using graph cuts
  • Hoch03: The Pseudoflow algorithm: A new algorithm for the maximum flow problem
  • Ishi03: Exact Optimization for Markov Random Fields with Convex Priors
  • Kolm03: Graph based algorithms for scene reconstruction from two or more views (dissertation)
  • Kolm03b: Generalized multi-camera scene reconstruction using graph cuts
  • Kovt03: Partial optimal labeling search for a NP-hard subclass of (max,+) problems
  • Kwat03: Graphcut Textures: Image and Video Synthesis Using Graph Cuts
  • Kim03: Visual Correspondence Using Energy Minimization and Mutual Information
  • Lin03: Surfaces with Occlusions from Layered Stereo
  • Pari03: A Volumetric Reconstruction Method from Multiple Calibrated Views using Global Graph Cut Optimization
  • Sand03: Polynomial Time Algorithms for Network Information Flow: one source, many sinks
  • Tapp03: Comparison of Graph Cuts with Belief Propagations for Stereo, using Identical MRF Parameters
  • Wang03: Image Segmentation with Ratio Cut
  • Xu03: Object Segmentation Using Graph Cuts Based Active Contours
  • Boyk02: An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision: the most popular algorithm.
  • Bueh02: Minimal Surfaces for stereo: tri-stereo in one min-cut.
  • Kolm02: Multi-camera Scene Reconstruction via Graph Cuts,
  • Roy02: Non-Uniform Hierarchical Pyramid Stereo for Large Images
  • Zale02: Fast algorithms of Bayesian Segmentation of Images "very fast special case" (Kolm03)
    
  • Boyk01: Interactive Graph Cuts for Optimal Boundary & Region Segmentation of Objects in NDImages
  • Boyk01b: A New Algorithm for Energy Minimization with Discontinuities
  • Iwat01: A Combinatorial Strongly Polynomial Algorithm for Minimizing Submodular Functions
  • Kang01: Handling occlusions in dense multi-view stereo
  • Kolm01: Computing Visual Correspondence with Occlusions via Graph Cuts: Seminal "expansion move" graph cut paper.
  • Iwat00: A Combinatorial Strongly Polynomial Algorithm for Minimizing Submodular Functions
  • Kim00: Incorporating spatial priors into an information theoretic approach for fMRI data analysis
  • Roy00: MRF solutions for probabilistic optical flow formulations
  • Shi00: Normalized Cuts and Image Segmentation.
  • Snow00: Exact Voxel Occumancy with Graph Cuts
  • Veks00: Image segmentation by nested cuts
  • Birc99: Multiway cut for stereo and motion with slanted surfaces: Alternately finding regions and their affine parameters.
  • Bobi99: Large Occlusion Stereo
  • Boyk99: Fast Approximate Energy Minimization via Graph Cuts ("expansion move", algorithm, beats jump move and swap move),
  • Boyk99b: A New Bayesian Framework for Object Recognition
  • Karg99: Random Sampling in Cut, Flow, and Network Design Problems
  • Klei99: Approximation Algorithms for Classification Problems with Pairwise Relationships: Metric Labeling and Markov Random Fields: "generalization of multiway cut"; "first polynomial-time algorithms".
  • Roy99: Stereo Without Epipolar Lines: A Maximum-Flow Formulation
  • Veks99: Efficient graph-based minimization methods in computer vision: "jump move" algorithm.
  • Ishi98: Occlusions, Discontinuities, and Epipolar Lines in Stereo
  • Ishi98b: Segmentation by Grouping Junctions one of the first image segmentation by max-flow papers.
  • Roy98: A maximum-flow formulation of the N-camera stereo correspondenceproblem : first use of graph cut for stereo reconstruction (Boyk02).
    
  • Chek97: Experimental Study of Minimum Cut Alorithms
  • Karg96: A New Approach to the Minimum Cut Problem: randomized algorithm
  • Hao94: A Faster Algorithm for Finding the Minimum Cut in a Directed Graph
  • Naga94: Implementing an efficient minimum capacity cut algorithm
  • Stoe94: A Simple Min-Cut Algorithm: One of the modern fast min-cut algorithms.
  • Wu93: An Optimal Graph Theoretic Approach to Data Clustering: Theory and Its Application to Image Segmentation
  • Gold88: A New Approach to the Maximum-Flow Problem: The first push-relabel algorithms.

• Graph algorithms other than min-cut:
  • Lafo06: Diffusion Maps and Coarse-Graining: A Unified Framework for Dimensionality Reduction, Graph Partitioning and Data Set Parameterization: clustering; nonlinear dimensionality reduction; markov random walks; reorganizing graphs.
  • Grad04c: Faster Graph-Theoretic Image Processing via Small-World and Quadtree Topologies
  • Hoch01: An efficient algorithm for image segmentation, Markov Random Fields and related problems: the case for convexity.

• Hashing techniques:
  • Gal06: Geometric Features for Partial Shape Matching and Similarity
  • Varm05: A Statistical Approach to Texture Classification from Single Images
  • Debo05: Scalable partitioning and exploration of chemical spaces using geometric hashing
  • Geor03: Mean Shift Based Clustering in High Dimensions: A Texture Classification Example
  • Shak03: Fast pose estimation with parameter-sensitive hashing
  • Gion99: Similarity Search in High Dimensions via Hashing: The seminal paper on Locality Sensitive Hashing.
  • Wolf97: Geometric Hashing: An Overview
  • Rigo97: Massively Parallel Bayesian Object Recognition: Geometric hashing on the connection machine. dissertation

• Histogram techniques:
  • Hou07: Multiview Pedestrian Detection Based on Vector Boosting
  • Ahon06: Face Description with Local Binary Patterns: Application to Face Recognition
  • Agar06b: Practical Global Optimization for Multiview Geometry
  • Dala06: Human Detection Using Oriented Histograms of Flow and Appearance
  • Grau06: Approximate Correspondences in High Dimensions
  • Lapt06: Improvements of Object Detection Using Boosted Histograms
  • Roth06: Cosegmentation of Image Pairs by Histogram Matching -- Incorporating a Global Constraint into MRFs
  • Schr06: Single-Histogram Class Models for Image Segmentation
  • Zhua06: Fast Human Detection Using a Cascade of Histograms of Oriented Gradients
  • Dala05: Histograms of Oriented Gradients for Human Detection
  • Wu05: Detection of Multiple, Partially Occluded Humans in a Single Image by Bayesian Combination of Edgelet Part Detectors
  • Ahon04: Face recognition with local binary patterns
  • Alle04: Object Tracking Using CamShift Algorithm and Multiple Quantized Feature Spaces
  • Cris04: A Multi-Stage Approach to Facial Feature Detection
  • Fei04: Joint Bayes Filter: A Hybrid Tracker for Non-Rigid Hand Motion Recognition
  • Laze04: A Sparse Texture Representation Using Local Affine Regions
  • Levi04: Learning Object Detection from a Small Number of Examples: the Importance of Good Features
  • Pati04: People Detection and Tracking in High Resolution Panoramic Video Mosaic
  • Roth04b: 3d Object Modelling and Recognition Using Local Affine-Invariant Image Descriptors and Multi-View Spatial Constraints: difference-of-gaussians (Dog); no segmentation state; handles high clutter levels; SIFT;
  • Liyu03: Foreground Object Detection from Videos Containing Complex Background
  • Thay02: Shape context and chamfer matching in cluttered scenes: Hand shapes.
  • Schn00: A Statistical Method for 3D Object Detection Applied to Faces and Cars
  • Swai91: Color indexing: "color histograms are stable over occlusion and change in view".
    

• Hidden Markov Models (HMM).
  • Crim06: Bilayer Segmentation of Live Video
  • Culo05: Gene prediction with conditional random fields
  • Smin05: Conditional models for contextual human motion recognition
  • Fei04: Joint Bayes Filter: A Hybrid Tracker for Non-Rigid Hand Motion Recognition
  • Pati04: People Detection and Tracking in High Resolution Panoramic Video Mosaic
  • Eddy03: Hidden Markov Models
  • Felz03: Fast Algorithms for Large-State-Space HMMs with Applications to Web Usage Analysis
  • Nevi99: Face Recognition Using an Embedded HMM
  • Smyt98: Belief Networks, Hidden Markov Models, and Markov Random Fields: a Unifying View
  • Rabi89: A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition

• Hough transform. (First popular robust method; essentially histogramming, canonically of line segments by approximate line membership.)
  • Leib07: Robust Object Detection with Interleaved Categorization and Segmentation
  • Stew99: Robust Parameter Estimation in Computer Vision.

• Image Analysis
  • Mogh08: Sparse Regression as a Sparse Eignvalue Problem
  • Zhou08: Multi-Instance Learning by Treating Instances As Non-I.I.D. Samples
  • Pata08: A Nonparametric Approach to 3D Shape Analysis from Digital Camera Images Mogh07: Fast Pixel/Part Selection with Sparse Eigenvectors: 1000X faster; eigenfaces; gender classification.
  • Nowo07: Weighted Substructure Mining for Image Analysis
  • He06: Learning and Incorporating Top-Down Cues in Image Segmentation
  • Russ06: Using multiple segmentations to discover objects and their extent in image collections
  • Yuil06: Vision as Bayesian inference: analysis by synthesis?
  • Roth05b: A framework for learning image priors
  • Wern05: A Linear Programming Approach to Max-Sum Problem: A Review
  • He04: Multiscale Conditional Random Fields for Image Labeling
  • Ho03: Clustering appearances of objects under varying illumination conditions

• Image Differencing.
  • Pati04: People Detection and Tracking in High Resolution Panoramic Video Mosaic
  • Rosi95: Image difference threshold strategies and shadow detection

• Image Matching and Restoration, Optical Flow, and Stereo Vision
  • Alah08: Reduce, Reuse & Recycle: Efficiently Solving Multilabel MRFs.
  • Kala08: The Five Points Pose Problem : A New and Accurate Solution Adapted to any Geometric Configuration
  • Sont08: Tightening LP Relaxations for MAP using message passing
  • Bhat07: Using Photographs to Enhance Videos of a Static Scene
  • Habb07: A Surface-Growing Approach to Multi-View Stereo Reconstruction
  • Hern07: Probabilistic visibility for multi-view stereo
  • Hirs07: Evaluation of Cost Functions for Stereo Matching: Rank and HMI (Hierarchical Mutual Information) win.
  • Lemp07: Global Optimization for Shape Fitting
  • Lemp07b: LogCut-Efficient Graph Cut Optimization for Markov Random Fields: "order of magnitude speedup"
  • Shar07: A Practical Approach for Super-Resolution using Photometric Stereo and Graph Cuts
  • Sinh07: Multi-View Stereo via Graph Cuts on the Dual of an Adaptive Tetrahedral Mesh
  • Vogi07: Multi-view Stereo via Volumetric Graph-cuts and Occlusion Robust Photo-Consistency
  • Wood07: On New View Synthesis Using Multiview Stereo
  • Wood07b: Efficient new-view synthesis using pairwise dictionary priors: Switching from large cliques to pairwise for speed.
  • Boyk06b: From Photohulls to Photoflux Optimization
  • Crem06: Statistical Priors for Efficient Combinatorial Optimization via Graph Cuts "hundreds of constraints per pixel"
  • Hirs06: Stereo Vision in Structured Environments by Consistent Semi-Global Matching
  • Pari06: A Surface Reconstruction Method Using Global Graph Cut Optimization
  • Yang06: Stereo Matching with Color-Weighted Correlation, Hierachical Belief Propagation and Occlusion Handling
  • Kolm06: Comparison of energy minimization algorithms for highly connected graphs
  • Kolm06b: Minimizing Nonsubmodular Functions with Graph Cuts -- A Review
  • Lemp06: Oriented visibility for multiview reconstruction
  • Mart06: View Synthesis for Multiview Video Compression
  • Roth06: Cosegmentation of Image Pairs by Histogram Matching -- Incorporating a Global Constraint into MRFs
  • Stic06: Global Depth from Epipolar Volumes--A General Framework for Reconstructing Non-Lambertian Surfaces "reconstruct highly specular surfaces"
    
  • Sun06: Using Strong Shape Priors for Stereo
  • Szel06: A Comparative Study of Energy Minimization Methods for Markov Random Fields
  • Szel06b: Image alignment and stitching: a tutorial
  • Will06: A Switched Gaussian Process for Estimating Disparity and Segmentation in Binocular Stereo
  • Wu06b: Dense Photometric Stereo: A Markov Random Field Approach: Beyond Lambertian surface models. Mirrored balls. :-)
  • Barb05b: Generalizing Swendsen-Wang to Sampling Arbitrary Posterior Probabilities
  • Chen05: Computer Vision Workload Analysis: Case Study of Video Surveillance Systems
  • Cheu05: Video Epitomes
  • Drou05: Geo-consistency for Wide Multi-Camera Stero
  • Hamp05: Smart Video Surveillance (magazine survey article)
  • Hirs05: Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information
  • Kolm05: Probabilistic fusion of stereo with color and contrast for bi-layer segmentation
  • Kolm05b: Convergent Tree-Reweighted Message Passing for Energy Minimization
  • Kolm05e: Bi-layer segmentation of binocular stereo video
  • Kolm05f: Primal-dual Algorithm for Convex Markov Random Fields: "for the panoramic stitching problem our method outperforms other techniques"
  • Melt05: Globally optimal solutions for energy minimization in stereo vision using reweighted belief propagation
  • Pape05: Highly Accurate Optic Flow Computation with Theoretically Justified Warping
  • Roth05: Digital Tapestry: Extending expansion move with non-metric hard and soft constraints.
  • Roth05b: A framework for learning image priors
  • Seit05: A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms
  • Sinh05: Multi-view Reconstruction using Photo-consistency and Exact Silhouette Constraints: A Maximum-Flow Formulation
  • Stew05: A Minimal Solution for Relative Pose with Unknown Focal Length
  • Vogi05: Multi-view Stereo via Volumetric Graph-cuts
  • Agar04: Interactive digital photomontage
  • Felz04: Efficient Belief Propagation for Early Vision
  • Hong04: Segment-based stereo matching using graph cuts: moving from pixel domain to segment domain
  • Mitt04: Motion-Based Background Subtraction Using Adaptive Kernel Density Estimation
  • Roth04b: 3d Object Modelling and Recognition Using Local Affine-Invariant Image Descriptors and Multi-View Spatial Constraints: RANSAC
  • Urta04: 3D Human Body Tracking using Deterministic Temporal Motion Models
  • Zitn04: High-quality video view interpolation using a layered representation
  • Dror03: Fragment-Based Image Completion
  • Kolm03b: Generalized multi-camera scene reconstruction using graph cuts
  • Jin03: Multi-view Stereo Beyond Lambert: Radiance tensors.
  • Lin03: Surfaces with Occlusions from Layered Stereo
  • Levi03: Learning How to Inpaint from Global Image Statistics: loopy belief propagaion.
  • Pari03: A Volumetric Reconstruction Method from Multiple Calibrated Views using Global Graph Cut Optimization
  • Tapp03: Comparison of Graph Cuts with Belief Propagations for Stereo, using Identical MRF Parameters
  • Kim03: Visual Correspondence Using Energy Minimization and Mutual Information
  • Kolm03: Graph based algorithms for scene reconstruction from two or more views (dissertation)
  • Kwat03: Graphcut Textures: Image and Video Synthesis Using Graph Cuts
  • Smin03: Kinematic jump processes for monocular 3D human tracking
  • Smin03b: Estimating Articulated Human Motion with Covariance Scaled Sampling
  • Alva02: Symmetrical Dense Optical Flow Estimation with Occlusion Detection
  • Bory02: An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Engergy Minimization in Vision,
  • Bueh02: Minimal Surfaces for stereo: tri-stereo in one min-cut.
  • Kolm02: Multi-camera Scene Reconstruction via Graph Cuts,
  • Roy02: Non-Uniform Hierarchical Pyramid Stereo for Large Images
  • Scha02: A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms
  • Sun02: Stereo Matching Using Belief Propagation
  • Szel02: Symmetric sub-pixel stereo matching
  • Leun01: Representing and Recognizing the Visual Appearance of Materials using Three-dimensional Textons
  • Kang01: Handling occlusions in dense multi-view stereo
  • Kolm01: Computing Visual Correspondence with Occlusions via Graph Cuts.
  • Smin01: Covariance scaled sampling for monocular 3D body tracking
  • Roy00: MRF solutions for probabilistic optical flow formulations
  • Shi00: Normalized Cuts and Image Segmentation.
  • Boug00: Pyramidal Implementation of the Lucas Kanade Feature Tracker: Description of the algorithm
  • Kutu00: A Theory of Shape by Space Carving
  • Vedu00: Shape and motion carving in 6 D
  • Birc99: Multiway cut for stereo and motion with slanted surfaces: Alternately finding regions and their affine parameters.
  • Bobi99: Large Occlusion Stereo
  • Boyk99b: A New Bayesian Framework for Object Recognition
  • Mihc99: Low-complexity image denoising based on statistical modeling of wavelet coefficients
  • Roy99: Stereo Without Epipolar Lines: A Maximum-Flow Formulation
  • Seit99: Photorealistic Scene Reconstruction by Voxel Coloring
  • Stew99: Robust Parameter Estimation in Computer Vision.
  • Szel99: An experimental comparison of stereo algorithms
  • Birc98: A Pixel Dissimilarity Measure That Is Insensitive to Image Sampling
  • Boyk98: Markov Random Fields with Efficient Approximations
  • Boyk98b: A Variable Window Approach to Early Vision
  • Ishi98: Occlusions, Discontinuities, and Epipolar Lines in Stereo
  • Roy98: A maximum-flow formulation of the N-camera stereo correspondenceproblem : first use of graph cut for stereo reconstruction (Boyk02).
    
  • Szel97: Creating full view panoramic image mosaics and environment maps
  • Horn81: Determining Optical Flow

• Image Segmentation and Background Models
  • Ding08: Two-Frames Accurate Motion Segmentation Using Tensor Voting and Graph Cuts
  • Khol08: Robust Higher Order Potentials for Enforcing Label Consistency.
  • Khol08b: On Partial Optimality in Multilabel MRFs.
  • Kuma08: Learning Layered Motion Segmentations of Video.
  • Camp07: Automatic 3D Object Segmentation in Multiple Views using Volumetric Graph-Cuts
  • Kim07: A new graph cut-based multiple active contour algorithm without initial contours and seed points
  • Kohl07: Minimizing Dynamic and Higher Order Energy Functions using Graph Cuts
  • Kolm07: Applications of parametric maxflow in computer vision
  • Leib07: Robust Object Detection with Interleaved Categorization and Segmentation
  • Rama07: Using Segmentation to Verify Object Hypotheses
  • Roth07: Optimizing Binary MRFs via Extended Roof Duality
  • Yin07: Tree-based Classifiers for Bilayer Video Segmentation
  • Appl06: Globally minimal surfaces by continuous maximal flows
  • Boyk06c: Graph Cuts and Efficient NDImage Segmentation
  • Bray06: PoseCut: Simultaneous Segmentation and 3D Pose Estimation of Humans using Dynamic Graph-Cuts.
  • Crem06: Kernel Density Estimation and Intrinsic Alignment for Shape Priors in Level Set Segmentation
  • Grub06: Incorporating non-motion cues into 3D motion segmentation
  • He06: Learning and Incorporating Top-Down Cues in Image Segmentation
  • Juan06: Active Graph Cuts: Speed-ups via good initial cut estimate.
  • Levi06: Learning to combine bottom-up and top-down segmentation
  • Rabi06: Model Order Selection and Cue Combination for Image Segmentation
  • Riha06: OBJCUT for face detection
  • Roth06: Cosegmentation of Image Pairs by Histogram Matching -- Incorporating a Global Constraint into MRFs
  • Russ06: Using multiple segmentations to discover objects and their extent in image collections
  • Scho06: Near real-time motion segmentation using graph cuts
  • Schr06: Single-Histogram Class Models for Image Segmentation
  • Shot06: Textonboost: Joint appearance, shape and context modeling for multi-class object recognition and segmentation
  • Barb05: Cluster Sampling and Its Applications in Image Analysis
  • Barb05b: Generalizing Swendsen-Wang to Sampling Arbitrary Posterior Probabilities
  • Drou05: Geo-consistency for Wide Multi-Camera Stero
  • Grad05: Multilabel Random Walker Image Segmentation Using Prior Models: "deep connection with graph cuts"
  • Hoie05: Automatic Photo Pop-up
  • Hoie05b: Geometric Context from a Single Image
  • Kohl05: Efficiently Solving Dynamic Markov Random Fields Using Graph Cuts
  • Kolm05g: Graph Cut Algorithms for Binocular Stereo with Occlusions
  • Plue05: New Active Contour Scheme for Cardiac MR Image Segmentation
  • Rive05: Entropy controlled Gauss-Markov random measure field models for early vision
  • Tian05: Robust and Efficient Foreground Analysis for Real-time Video Surveillance
  • Tu05: Image Parsing: Unifying Segmentation, Detection, and Recognition
  • Blak04: Interactive image segmentation using an adapative GMMRF model
  • Cheu04: Robust techniques for background subtraction in urban traffic video
  • Crem04: Kernel Density Estimation and Intrinsic Alignment for Knowledge-driven Segmentation: Teaching Level Sets to Walk
  • Grad04b: Multi-Label Image Segmentation for Medical Applications Based on Graph-Theoretic Electrical Potentials
  • Grad04c: Faster Graph-Theoretic Image Processing via Small-World and Quadtree Topologies
  • Howe04: Better Foreground Segmentation Through Graph Cuts
  • Mitt04: Motion-Based Background Subtraction Using Adaptive Kernel Density Estimation
  • Roth04: "GrabCut": interactive foreground extraction using iterated graph cuts
  • Zabi04: Spatially coherent clustering using graph cuts: "operates simultaneously in feature and image space"
  • Boyk03: Computing geodesics and minimal surfaces via graph cuts
  • Gold03: Joint 3D-reconstruction and background separation in multiple views using graph cuts
  • Liyu03: Foreground Object Detection from Videos Containing Complex Background
  • Pori03: Human Body Tracking by Adaptive Background Models and Mean-Shift Analysis
  • Ren03: Learning a classification model for segmentation
  • Wang03: Image Segmentation with Ratio Cut
  • Xu03: Object Segmentation Using Graph Cuts Base Active Contours
  • Boyk02: An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Engergy Minimization in Vision,
  • Coma02: Mean Shift: A Robust Approach toward Feature Space Analysis
  • Tu02: Image Segmentation by Data-Driven Markov Chain Monte Carlo
  • Zale02: Fast algorithms of Bayesian Segmentation of Images "very fast special case" (Kolm03)
    
  • Zhu02: What are Textons?
  • Boyk01: Interactive Graph Cuts for Optimal Boundary & Region Segmentation of Objects in NDImages
  • Hoch01: An efficient algorithm for image segmentation, Markov Random Fields and related problems.
  • Kade01: An Improved Adaptive Background Mixture Model for Real-Time Tracking with Shadow Detection: The mixture-of-gaussians (mog) paper which is the basis for the OpenCV blobtracker mog module.
  • Kim00: Incorporating spatial priors into an information theoretic approach for fMRI data analysis
  • Shi00: Normalized Cuts and Image Segmentation.
  • Side00b: Stochastic tracking of 3D human figures using 2D image motion
  • Stau00: Learning Patterns of Activity Using Real-Time Tracking: Reference example of implementing mixture-of-gaussians (mog) background subtraction.
  • Veks00: Image segmentation by nested cuts
  • Elga99: Non-parametric Model for Background Subraction: Kernel methods, shadow suppression. Generalization of mixture-of-gaussians (mog)
  • Stau99: Adaptive background mixture models for real-time tracking
  • Hofm98: Unsupervised Texture Segmentation in a Deterministic Annealing Framework
  • Ishi98b: Segmentation by Grouping Junctions one of the first image segmentation by max-flow papers.
  • Frie97: Image segmentation in video sequences: A probabilistic approach (-the- seminal mixture-of-gaussians background model paper)
  • Wren97: Pfinder: Real-Time Tracking of the Human Body (a seminal mixture-of-gaussians background model paper)
  • Boum96: A Multiscale Random Field Model for Bayesian Image Segmentation
  • Wu93: An Optimal Graph Theoretic Approach to Data Clustering: Theory and Its Application to Image Segmentation

• Isoperimetric Algorithms.
  • Grad04: Isoperimetric Graph Partitioning for Data Clustering and Image Segmentation
  • Grad04b: Multi-Label Image Segmentation for Medical Applications Based on Graph-Theoretic Electrical Potentials
  • Grad04c: Faster Graph-Theoretic Image Processing via Small-World and Quadtree Topologies

• Kuma03: Discriminative Random Fields: A Discriminative Framework for Contextual Interaction in Classification: adding label neighborhood interations.