Graph Cuts. Srikumar Ramalingam School of Computing University of Utah

Size: px
Start display at page:

Download "Graph Cuts. Srikumar Ramalingam School of Computing University of Utah"

Transcription

1 Graph Cuts Srikumar Ramalingam School o Computing University o Utah

2 Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion

3 Segmentation Problem [Boykov and Jolly 00, Rother et al. 004]

4 Stereo Reconstruction Choose the disparities rom the discrete set: (,,..., L)

5 Image Denoising Original Denoised image

6 Semantic Labeling (Building, ground, sky) [Hoiem, Eros, Hebert, IJCV, 007 ]

7 Image Labeling Problems

8 Labeling is highly structured Highly unlikely Image Courtesy: Lubor Ladicky

9 Labeling is highly structured Slide Courtesy: Lubor Ladicky

10 Image Labeling Problems Slide Courtesy: Lubor Ladicky

11 Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion

12 Pseudo Boolean Functions (PBF) Variables: x, x,..., 0, x n Negations: 0, xi x i Pseudo-Boolean Functions (PBF): :{0,} n» Maps a Boolean vector to a real number. R Has unique multi-linear representation:» For example: ( x x x x, x, x3, x4) 3xx4 5 3 [Boros&Hammer 00]

13 Posiorms or Pseudo-Boolean unctions (PBF) Posiorms: Non-negative multi-linear polynomial except maybe the constant terms. ( x x x x, x, x3, x4) 3xx4 5 3( x ) x x x x 4 5 3x 3x x x 4 4 5xx 3 x4) 3xx4 5x x4 3xx4 5xx 3( x x 3 x Several posiorms exist or a given unction. Provides bounds or minimization, e.g [Boros&Hammer 00] 3

14 Set Functions are Pseudo Boolean Functions (PBF) Finite ground set V {,,..., n} Set unction (Input - subset o, output - real number) s : V V R - correspondence exists between and subset V S {,,3,4} o V. { x, x, x3 0, x4 } x (,,4), x,..., x n x x i i 0 0, i i S S

15 Set Functions are Pseudo Boolean Functions (PBF) Consider a PBF ( x, x, x3, x4) 3xx4 5xx3 Equivalent to a set unction s s ({,}) 3()(0) 5()(0) ({,3}) 3()(0) 5()() 7

16 Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion

17 Submodular set unctions (Union- Intersection) A set unction is submodular i and only i: V B A B A B A B A, ), ( ) ( ) ( ) ( R V : V B A,

18 Equivalent Deinitions Slide Courtesy: Krause, Jegelka

19 Questions Slide Courtesy: Krause, Jegelka

20 Submodularity Example Slide Courtesy: Krause, Jegelka

21 Submodularity Example Slide Courtesy: Krause, Jegelka

22 Submodularity Example Slide Courtesy: Krause, Jegelka

23 Submodularity Example Slide Courtesy: Krause, Jegelka

24 Set cover is submodular Slide Courtesy: Krause, Jegelka

25 Submodular set unctions (Union- Intersection) ( A) ( B) ( A B) ( A B), A, B V Let us consider a very simple case with only two variables x and x. V {,}, A {}, B {} (0,0) (0,) Using submodularity, we have: ( x ( x, x, x 0) ) ( x ( x ) 0) (,0) (0,) (,) (0,0) 0, x 0, x (,0) (, ) Main diagonal elements are smaller than o-diagonal ones. Blue is larger than red.

26 Quadratic Pseudo Boolean Functions (QPBF) Example o quadratic pseudo Boolean unctions ( x, x, x, x ) x 3x x x x x [Boros&Hammer 00]

27 Submodular Quadratic Pseudo Boolean Functions A QPBF is submodular i and only i all quadratic coeicients are non-positive. ( x, x, x ) 5 x 3x x x x x

28 Example or submodular QPBF ( x, x, x ) 5 x 3x 3x x x x V {,,3}, A {,}, B {,3} A B {,,3}, A B {} ( A) 5 3() 3()(0) 5()(0) 3 3 ( B) 5 0 3() 3(0)() 5()() 7 ( A B) 5 3() 3()() 5()() 5 ( A B) 5 0 3() 3(0)(0) 5()(0) ( A) ( B) ( A B) ( A B),(3 7 5 )

29 Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion

30 Max-low/Min-cut Image courtesy: Lubor Ladicky

31 Max-low/Min-cut

32 Max-low/Min-cut

33 Max-low/Min-cut

34 Network model or submodular QPBF A submodular QPBF can be associated with a network. There is - correspondence every edge in network and every term in. Let us denote source by s 0 and sink by t. An edge that goes rom to x is denoted by x x. x G v x x s x x x sx x t x t

35 Network model or submodular QPBF x x x x s x sx x x x xt t Given a QPBF we rewrite it using a posiorm representation using only three types o terms: x, x,, i x j i x i 3x x x x 4 3x x ( 4xx 4x 4x) 3x x 4( x ) x x 4 3 s 3x x x 3x 4 x 4 x 3x ( 3x 3 3) x x 4 3 3x x x 3( x) 4 3 3sx 3x t x x 4 t 3

36 Network model or submodular QPBF There is a one-one correspondence between values o and s-t cut values o [Hammer 965] 3 sx s 4 G v. x x x 3 x t t x ( x C({ x 0, x, x ) }) 4 s-t mincut [Ford&Fulkerson 6, Goldberg&Tarzan86]

37 Network model or submodular QPBF There is a one-one correspondence between values o and s-t cut values o [Hammer 965]. s 3 4 G v x 3 x Thus we can compute the minimum o algorithm on the associated G v. t ( x C({ x, x, s},{ x using maxlow/mincut 0), t}) s-t mincut [Ford&Fulkerson 6, Goldberg&Tarzan86]

38 Network model or non-submodular QPBF A non-submodular QBPF can be associated with a network as ollows: 3x x 4x x x G v s x 3x 3sx 5x 5sx 4( x 4x x ) x t There is no polynomial-time algorithm or s-t mincut on a network with negative edge capacities. A submodular QBPF can always be associated with a network with non-negative edge capacities.

39 Minimizing Quadratic Pseudo Boolean Functions I QPBF is submodular, use maxlow algo.. [Ford&Fulkerson 6, Goldberg&Tarzan86] I QPBF is non-submodular, Belie propagation or other message passing algorithms. [Boros&Hammer 00]

40 Multi-label Problems Choose the disparities rom the discrete set: (,,..., L)

41 Multi-label Problems Exact Methods: Transorm the given multi-label problems to Boolean problems and solve them using maxlow/mincut algorithms or QPBO techniques. [Not covered in this course!] Approximate Methods: Develop iterative move-making algorithms where each move corresponds to a Boolean problem.

42 Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion

43 Boolean Energy Function Variables x, x,..., 0,. x n j - cost o assigning x i j {0, }. x i lm x x i j x i l x j m. - cost o jointly assigning and Energy unction: E( x j j j j ij i j, x) x x x x xx x x j0 j0 i0 j0

44 Energy Move Making Algorithms Solution Space [Image courtesy: Pushmeet Kohli, Phil Torr]

45 Energy Move Making Algorithms Current Solution Search Neighbourhood Optimal Move Solution Space [Image courtesy: Pushmeet Kohli, Phil Torr]

46 Expansion building [Boykov et al. 00]

47 Expansion Let yi and y j be two adjacent variables whose labels are not. retain y i l a y j retain lb In the move space, we compute i the two variables should retain the same labels or move to label. [Boykov et al. 00]

48 Expansion In the move space, we use two Boolean variables iand jto denote and y respectively. The encoding is shown below: y i y y i i l a j x i 0 x y x i Submodularity condition states that the sum o main diagonal elements is less than the sum o elements in the o-diagonal: y j j l a x j j 0 x x 00 x i x j 0 x i x j 0 x i x j x i x j = l y a i i l y b lb y y j j a l y y i y i y j j [Boykov et al. 00]

49 Submodularity condition states that the sum o main diagonal elements is less than the sum o elements in the o-diagonal:

50 Expansion [Image courtesy: Lubor Ladicky] [Boykov et al. 00]

51 Thank You

Graph Cuts. Srikumar Ramalingam School of Computing University of Utah

Graph Cuts. Srikumar Ramalingam School of Computing University of Utah Graph Cuts Srikumar Ramalingam School o Computing University o Utah Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion Segmentation Problem [Boykov

More information

Models for grids. Computer vision: models, learning and inference. Multi label Denoising. Binary Denoising. Denoising Goal.

Models for grids. Computer vision: models, learning and inference. Multi label Denoising. Binary Denoising. Denoising Goal. Models for grids Computer vision: models, learning and inference Chapter 9 Graphical Models Consider models where one unknown world state at each pixel in the image takes the form of a grid. Loops in the

More information

Quadratic Pseudo-Boolean Optimization(QPBO): Theory and Applications At-a-Glance

Quadratic Pseudo-Boolean Optimization(QPBO): Theory and Applications At-a-Glance Quadratic Pseudo-Boolean Optimization(QPBO): Theory and Applications At-a-Glance Presented By: Ahmad Al-Kabbany Under the Supervision of: Prof.Eric Dubois 12 June 2012 Outline Introduction The limitations

More information

Automatic Dense Semantic Mapping From Visual Street-level Imagery

Automatic Dense Semantic Mapping From Visual Street-level Imagery Automatic Dense Semantic Mapping From Visual Street-level Imagery Sunando Sengupta [1], Paul Sturgess [1], Lubor Ladicky [2], Phillip H.S. Torr [1] [1] Oxford Brookes University [2] Visual Geometry Group,

More information

Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction

Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction Marc Pollefeys Joined work with Nikolay Savinov, Christian Haene, Lubor Ladicky 2 Comparison to Volumetric Fusion Higher-order ray

More information

Computer Vision : Exercise 4 Labelling Problems

Computer Vision : Exercise 4 Labelling Problems Computer Vision Exercise 4 Labelling Problems 13/01/2014 Computer Vision : Exercise 4 Labelling Problems Outline 1. Energy Minimization (example segmentation) 2. Iterated Conditional Modes 3. Dynamic Programming

More information

Markov/Conditional Random Fields, Graph Cut, and applications in Computer Vision

Markov/Conditional Random Fields, Graph Cut, and applications in Computer Vision Markov/Conditional Random Fields, Graph Cut, and applications in Computer Vision Fuxin Li Slides and materials from Le Song, Tucker Hermans, Pawan Kumar, Carsten Rother, Peter Orchard, and others Recap:

More information

A Principled Deep Random Field Model for Image Segmentation

A Principled Deep Random Field Model for Image Segmentation A Principled Deep Random Field Model for Image Segmentation Pushmeet Kohli Microsoft Research Cambridge, UK Anton Osokin Moscow State University Moscow, Russia Stefanie Jegelka UC Berkeley Berkeley, CA,

More information

Image Segmentation. Srikumar Ramalingam School of Computing University of Utah. Slides borrowed from Ross Whitaker

Image Segmentation. Srikumar Ramalingam School of Computing University of Utah. Slides borrowed from Ross Whitaker Image Segmentation Srikumar Ramalingam School of Computing University of Utah Slides borrowed from Ross Whitaker Segmentation Semantic Segmentation Indoor layout estimation What is Segmentation? Partitioning

More information

Markov Random Fields and Segmentation with Graph Cuts

Markov Random Fields and Segmentation with Graph Cuts Markov Random Fields and Segmentation with Graph Cuts Computer Vision Jia-Bin Huang, Virginia Tech Many slides from D. Hoiem Administrative stuffs Final project Proposal due Oct 27 (Thursday) HW 4 is out

More information

MRFs and Segmentation with Graph Cuts

MRFs and Segmentation with Graph Cuts 02/24/10 MRFs and Segmentation with Graph Cuts Computer Vision CS 543 / ECE 549 University of Illinois Derek Hoiem Today s class Finish up EM MRFs w ij i Segmentation with Graph Cuts j EM Algorithm: Recap

More information

CAP5415-Computer Vision Lecture 13-Image/Video Segmentation Part II. Dr. Ulas Bagci

CAP5415-Computer Vision Lecture 13-Image/Video Segmentation Part II. Dr. Ulas Bagci CAP-Computer Vision Lecture -Image/Video Segmentation Part II Dr. Ulas Bagci bagci@ucf.edu Labeling & Segmentation Labeling is a common way for modeling various computer vision problems (e.g. optical flow,

More information

CS6670: Computer Vision

CS6670: Computer Vision CS6670: Computer Vision Noah Snavely Lecture 19: Graph Cuts source S sink T Readings Szeliski, Chapter 11.2 11.5 Stereo results with window search problems in areas of uniform texture Window-based matching

More information

Learning CRFs using Graph Cuts

Learning CRFs using Graph Cuts Appears in European Conference on Computer Vision (ECCV) 2008 Learning CRFs using Graph Cuts Martin Szummer 1, Pushmeet Kohli 1, and Derek Hoiem 2 1 Microsoft Research, Cambridge CB3 0FB, United Kingdom

More information

GEOMETRIC SEGMENTATION USING GAUSSIAN WAVE PACKETS

GEOMETRIC SEGMENTATION USING GAUSSIAN WAVE PACKETS Proceedings of the Project Review, Geo-Mathematical Imaging Group (Purdue University, West Lafayette IN), Vol. 1 (2011) pp. 57-64. GEOMETRIC SEGMENTATION USING GAUSSIAN WAVE PACKETS FREDRIK ANDERSSON AND

More information

Discrete Optimization Methods in Computer Vision CSE 6389 Slides by: Boykov Modified and Presented by: Mostafa Parchami Basic overview of graph cuts

Discrete Optimization Methods in Computer Vision CSE 6389 Slides by: Boykov Modified and Presented by: Mostafa Parchami Basic overview of graph cuts Discrete Optimization Methods in Computer Vision CSE 6389 Slides by: Boykov Modified and Presented by: Mostafa Parchami Basic overview of graph cuts [Yuri Boykov, Olga Veksler, Ramin Zabih, Fast Approximation

More information

Reduce, Reuse & Recycle: Efficiently Solving Multi-Label MRFs

Reduce, Reuse & Recycle: Efficiently Solving Multi-Label MRFs Reduce, Reuse & Recycle: Efficiently Solving Multi-Label MRFs Karteek Alahari 1 Pushmeet Kohli 2 Philip H. S. Torr 1 1 Oxford Brookes University 2 Microsoft Research, Cambridge Abstract In this paper,

More information

Introduction à la vision artificielle X

Introduction à la vision artificielle X Introduction à la vision artificielle X Jean Ponce Email: ponce@di.ens.fr Web: http://www.di.ens.fr/~ponce Planches après les cours sur : http://www.di.ens.fr/~ponce/introvis/lect10.pptx http://www.di.ens.fr/~ponce/introvis/lect10.pdf

More information

Big and Tall: Large Margin Learning with High Order Losses

Big and Tall: Large Margin Learning with High Order Losses Big and Tall: Large Margin Learning with High Order Losses Daniel Tarlow University of Toronto dtarlow@cs.toronto.edu Richard Zemel University of Toronto zemel@cs.toronto.edu Abstract Graphical models

More information

Approximate MRF Inference Using Bounded Treewidth Subgraphs

Approximate MRF Inference Using Bounded Treewidth Subgraphs Approximate MRF Inference Using Bounded Treewidth Subgraphs Alexander Fix 1, Joyce Chen 1, Endre Boros 2, and Ramin Zabih 1 1 Cornell University, Computer Science Department, Ithaca, New York {afix,yuhsin,rdz}@cs.cornell.edu

More information

Multi-Label Moves for Multi-Label Energies

Multi-Label Moves for Multi-Label Energies Multi-Label Moves for Multi-Label Energies Olga Veksler University of Western Ontario some work is joint with Olivier Juan, Xiaoqing Liu, Yu Liu Outline Review multi-label optimization with graph cuts

More information

MULTI-REGION SEGMENTATION

MULTI-REGION SEGMENTATION MULTI-REGION SEGMENTATION USING GRAPH-CUTS Johannes Ulén Abstract This project deals with multi-region segmenation using graph-cuts and is mainly based on a paper by Delong and Boykov [1]. The difference

More information

Energy Minimization Under Constraints on Label Counts

Energy Minimization Under Constraints on Label Counts Energy Minimization Under Constraints on Label Counts Yongsub Lim 1, Kyomin Jung 1, and Pushmeet Kohli 2 1 Korea Advanced Istitute of Science and Technology, Daejeon, Korea yongsub@kaist.ac.kr, kyomin@kaist.edu

More information

In Defense of 3D-Label Stereo

In Defense of 3D-Label Stereo 2013 IEEE Conference on Computer Vision and Pattern Recognition In Defense of 3D-Label Stereo Carl Olsson Johannes Ulén Centre for Mathematical Sciences Lund University, Sweden calle@maths.lth.se ulen@maths.lth.se

More information

Learning Low-order Models for Enforcing High-order Statistics

Learning Low-order Models for Enforcing High-order Statistics Patrick Pletscher ETH Zurich Zurich, Switzerland Pushmeet Kohli Microsoft Research Cambridge, UK Abstract Models such as pairwise conditional random fields (CRFs) are extremely popular in computer vision

More information

Robust Higher Order Potentials for Enforcing Label Consistency

Robust Higher Order Potentials for Enforcing Label Consistency Int J Comput Vis (2009) 82: 302 324 DOI 10.1007/s11263-008-0202-0 Robust Higher Order Potentials for Enforcing Label Consistency Pushmeet Kohli L ubor Ladický Philip H.S. Torr Received: 17 August 2008

More information

GRMA: Generalized Range Move Algorithms for the Efficient Optimization of MRFs

GRMA: Generalized Range Move Algorithms for the Efficient Optimization of MRFs International Journal of Computer Vision manuscript No. (will be inserted by the editor) GRMA: Generalized Range Move Algorithms for the Efficient Optimization of MRFs Kangwei Liu Junge Zhang Peipei Yang

More information

Introduction to Computer Vision. Srikumar Ramalingam School of Computing University of Utah

Introduction to Computer Vision. Srikumar Ramalingam School of Computing University of Utah Introduction to Computer Vision Srikumar Ramalingam School of Computing University of Utah srikumar@cs.utah.edu Course Website http://www.eng.utah.edu/~cs6320/ What is computer vision? Light source 3D

More information

Markov Networks in Computer Vision

Markov Networks in Computer Vision Markov Networks in Computer Vision Sargur Srihari srihari@cedar.buffalo.edu 1 Markov Networks for Computer Vision Some applications: 1. Image segmentation 2. Removal of blur/noise 3. Stereo reconstruction

More information

Segmentation. Bottom up Segmentation Semantic Segmentation

Segmentation. Bottom up Segmentation Semantic Segmentation Segmentation Bottom up Segmentation Semantic Segmentation Semantic Labeling of Street Scenes Ground Truth Labels 11 classes, almost all occur simultaneously, large changes in viewpoint, scale sky, road,

More information

Alina Ene. From Minimum Cut to Submodular Minimization Leveraging the Decomposable Structure. Boston University

Alina Ene. From Minimum Cut to Submodular Minimization Leveraging the Decomposable Structure. Boston University From Minimum Cut to Submodular Minimization Leveraging the Decomposable Structure Alina Ene Boston University Joint work with Huy Nguyen (Northeastern University) Laszlo Vegh (London School of Economics)

More information

W[1]-hardness. Dániel Marx. Recent Advances in Parameterized Complexity Tel Aviv, Israel, December 3, 2017

W[1]-hardness. Dániel Marx. Recent Advances in Parameterized Complexity Tel Aviv, Israel, December 3, 2017 1 W[1]-hardness Dániel Marx Recent Advances in Parameterized Complexity Tel Aviv, Israel, December 3, 2017 2 Lower bounds So far we have seen positive results: basic algorithmic techniques for fixed-parameter

More information

Markov Networks in Computer Vision. Sargur Srihari

Markov Networks in Computer Vision. Sargur Srihari Markov Networks in Computer Vision Sargur srihari@cedar.buffalo.edu 1 Markov Networks for Computer Vision Important application area for MNs 1. Image segmentation 2. Removal of blur/noise 3. Stereo reconstruction

More information

Markov Random Fields and Segmentation with Graph Cuts

Markov Random Fields and Segmentation with Graph Cuts Markov Random Fields and Segmentation with Graph Cuts Computer Vision Jia-Bin Huang, Virginia Tech Many slides from D. Hoiem Administrative stuffs Final project Proposal due Oct 30 (Monday) HW 4 is out

More information

Segmentation in electron microscopy images

Segmentation in electron microscopy images Segmentation in electron microscopy images Aurelien Lucchi, Kevin Smith, Yunpeng Li Bohumil Maco, Graham Knott, Pascal Fua. http://cvlab.epfl.ch/research/medical/neurons/ Outline Automated Approach to

More information

UNIT #2 TRANSFORMATIONS OF FUNCTIONS

UNIT #2 TRANSFORMATIONS OF FUNCTIONS Name: Date: UNIT # TRANSFORMATIONS OF FUNCTIONS Part I Questions. The quadratic unction ollowing does,, () has a turning point at have a turning point? 7, 3, 5 5, 8. I g 7 3, then at which o the The structure

More information

W[1]-hardness. Dániel Marx 1. Hungarian Academy of Sciences (MTA SZTAKI) Budapest, Hungary

W[1]-hardness. Dániel Marx 1. Hungarian Academy of Sciences (MTA SZTAKI) Budapest, Hungary W[1]-hardness Dániel Marx 1 1 Institute for Computer Science and Control, Hungarian Academy of Sciences (MTA SZTAKI) Budapest, Hungary School on Parameterized Algorithms and Complexity Będlewo, Poland

More information

USING ONE GRAPH-CUT TO FUSE MULTIPLE CANDIDATE MAPS IN DEPTH ESTIMATION

USING ONE GRAPH-CUT TO FUSE MULTIPLE CANDIDATE MAPS IN DEPTH ESTIMATION USING ONE GRAPH-CUT TO FUSE MULTIPLE CANDIDATE MAPS IN DEPTH ESTIMATION F. Pitié Sigmedia Group, Trinity College Dublin, fpitie@mee.tcd.ie Abstract Graph-cut techniques for depth and disparity estimations

More information

Image Segmentation with a Bounding Box Prior Victor Lempitsky, Pushmeet Kohli, Carsten Rother, Toby Sharp Microsoft Research Cambridge

Image Segmentation with a Bounding Box Prior Victor Lempitsky, Pushmeet Kohli, Carsten Rother, Toby Sharp Microsoft Research Cambridge Image Segmentation with a Bounding Box Prior Victor Lempitsky, Pushmeet Kohli, Carsten Rother, Toby Sharp Microsoft Research Cambridge Dylan Rhodes and Jasper Lin 1 Presentation Overview Segmentation problem

More information

Supplementary Material for ECCV 2012 Paper: Extracting 3D Scene-consistent Object Proposals and Depth from Stereo Images

Supplementary Material for ECCV 2012 Paper: Extracting 3D Scene-consistent Object Proposals and Depth from Stereo Images Supplementary Material for ECCV 2012 Paper: Extracting 3D Scene-consistent Object Proposals and Depth from Stereo Images Michael Bleyer 1, Christoph Rhemann 1,2, and Carsten Rother 2 1 Vienna University

More information

Dynamic Planar-Cuts: Efficient Computation of Min-Marginals for Outer-Planar Models

Dynamic Planar-Cuts: Efficient Computation of Min-Marginals for Outer-Planar Models Dynamic Planar-Cuts: Efficient Computation of Min-Marginals for Outer-Planar Models Dhruv Batra 1 1 Carnegie Mellon University www.ece.cmu.edu/ dbatra Tsuhan Chen 2,1 2 Cornell University tsuhan@ece.cornell.edu

More information

NP-complete Reductions

NP-complete Reductions NP-complete Reductions 1. Prove that 3SAT P DOUBLE-SAT, i.e., show DOUBLE-SAT is NP-complete by reduction from 3SAT. The 3-SAT problem consists of a conjunction of clauses over n Boolean variables, where

More information

Energy Minimization for Segmentation in Computer Vision

Energy Minimization for Segmentation in Computer Vision S * = arg S min E(S) Energy Minimization for Segmentation in Computer Vision Meng Tang, Dmitrii Marin, Ismail Ben Ayed, Yuri Boykov Outline Clustering/segmentation methods K-means, GrabCut, Normalized

More information

3 No-Wait Job Shops with Variable Processing Times

3 No-Wait Job Shops with Variable Processing Times 3 No-Wait Job Shops with Variable Processing Times In this chapter we assume that, on top of the classical no-wait job shop setting, we are given a set of processing times for each operation. We may select

More information

Analysis: TextonBoost and Semantic Texton Forests. Daniel Munoz Februrary 9, 2009

Analysis: TextonBoost and Semantic Texton Forests. Daniel Munoz Februrary 9, 2009 Analysis: TextonBoost and Semantic Texton Forests Daniel Munoz 16-721 Februrary 9, 2009 Papers [shotton-eccv-06] J. Shotton, J. Winn, C. Rother, A. Criminisi, TextonBoost: Joint Appearance, Shape and Context

More information

Lecture 16 Segmentation and Scene understanding

Lecture 16 Segmentation and Scene understanding Lecture 16 Segmentation and Scene understanding Introduction! Mean-shift! Graph-based segmentation! Top-down segmentation! Silvio Savarese Lecture 15 -! 3-Mar-14 Segmentation Silvio Savarese Lecture 15

More information

Measuring Uncertainty in Graph Cut Solutions - Efficiently Computing Min-marginal Energies using Dynamic Graph Cuts

Measuring Uncertainty in Graph Cut Solutions - Efficiently Computing Min-marginal Energies using Dynamic Graph Cuts Measuring Uncertainty in Graph Cut Solutions - Efficiently Computing Min-marginal Energies using Dynamic Graph Cuts Pushmeet Kohli Philip H.S. Torr Department of Computing, Oxford Brookes University, Oxford.

More information

Simultaneous Multi-class Pixel Labeling over Coherent Image Sets

Simultaneous Multi-class Pixel Labeling over Coherent Image Sets Simultaneous Multi-class Pixel Labeling over Coherent Image Sets Paul Rivera Research School of Computer Science Australian National University Canberra, ACT 0200 Stephen Gould Research School of Computer

More information

Lecture 1 Aug. 26, 2015

Lecture 1 Aug. 26, 2015 CS 388R: Randomized Algorithms Fall 015 Prof. Eric Price Lecture 1 Aug. 6, 015 Scribe: Ashish Bora, Jessica Hoffmann 1 Overview In this lecture, we get a first taste of randomized algorithms, from definition

More information

MATRIX ALGORITHM OF SOLVING GRAPH CUTTING PROBLEM

MATRIX ALGORITHM OF SOLVING GRAPH CUTTING PROBLEM UDC 681.3.06 MATRIX ALGORITHM OF SOLVING GRAPH CUTTING PROBLEM V.K. Pogrebnoy TPU Institute «Cybernetic centre» E-mail: vk@ad.cctpu.edu.ru Matrix algorithm o solving graph cutting problem has been suggested.

More information

Uncertainty Driven Multi-Scale Optimization

Uncertainty Driven Multi-Scale Optimization Uncertainty Driven Multi-Scale Optimization Pushmeet Kohli 1 Victor Lempitsky 2 Carsten Rother 1 1 Microsoft Research Cambridge 2 University of Oxford Abstract. This paper proposes a new multi-scale energy

More information

Image Segmentation with A Bounding Box Prior

Image Segmentation with A Bounding Box Prior Image Segmentation with A Bounding Box Prior Victor Lempitsky, Pushmeet Kohli, Carsten Rother, Toby Sharp Microsoft Research Cambridge Abstract User-provided object bounding box is a simple and popular

More information

Graphs and Network Flows IE411. Lecture 13. Dr. Ted Ralphs

Graphs and Network Flows IE411. Lecture 13. Dr. Ted Ralphs Graphs and Network Flows IE411 Lecture 13 Dr. Ted Ralphs IE411 Lecture 13 1 References for Today s Lecture IE411 Lecture 13 2 References for Today s Lecture Required reading Sections 21.1 21.2 References

More information

Graphs, graph algorithms (for image segmentation),... in progress

Graphs, graph algorithms (for image segmentation),... in progress Graphs, graph algorithms (for image segmentation),... in progress Václav Hlaváč Czech Technical University in Prague Czech Institute of Informatics, Robotics and Cybernetics 66 36 Prague 6, Jugoslávských

More information

ESSP: An Efficient Approach to Minimizing Dense and Nonsubmodular Energy Functions

ESSP: An Efficient Approach to Minimizing Dense and Nonsubmodular Energy Functions 1 ESSP: An Efficient Approach to Minimizing Dense and Nonsubmodular Functions Wei Feng, Member, IEEE, Jiaya Jia, Senior Member, IEEE, and Zhi-Qiang Liu, arxiv:145.4583v1 [cs.cv] 19 May 214 Abstract Many

More information

(Lec 14) Placement & Partitioning: Part III

(Lec 14) Placement & Partitioning: Part III Page (Lec ) Placement & Partitioning: Part III What you know That there are big placement styles: iterative, recursive, direct Placement via iterative improvement using simulated annealing Recursive-style

More information

CS599: Convex and Combinatorial Optimization Fall 2013 Lecture 14: Combinatorial Problems as Linear Programs I. Instructor: Shaddin Dughmi

CS599: Convex and Combinatorial Optimization Fall 2013 Lecture 14: Combinatorial Problems as Linear Programs I. Instructor: Shaddin Dughmi CS599: Convex and Combinatorial Optimization Fall 2013 Lecture 14: Combinatorial Problems as Linear Programs I Instructor: Shaddin Dughmi Announcements Posted solutions to HW1 Today: Combinatorial problems

More information

Image Segmentation. Marc Pollefeys. ETH Zurich. Slide credits: V. Ferrari, K. Grauman, B. Leibe, S. Lazebnik, S. Seitz,Y Boykov, W. Freeman, P.

Image Segmentation. Marc Pollefeys. ETH Zurich. Slide credits: V. Ferrari, K. Grauman, B. Leibe, S. Lazebnik, S. Seitz,Y Boykov, W. Freeman, P. Image Segmentation Perceptual and Sensory Augmented Computing Computer Vision WS 0/09 Marc Pollefeys ETH Zurich Slide credits: V. Ferrari, K. Grauman, B. Leibe, S. Lazebnik, S. Seitz,Y Boykov, W. Freeman,

More information

Solving problems on graph algorithms

Solving problems on graph algorithms Solving problems on graph algorithms Workshop Organized by: ACM Unit, Indian Statistical Institute, Kolkata. Tutorial-3 Date: 06.07.2017 Let G = (V, E) be an undirected graph. For a vertex v V, G {v} is

More information

CS 5540 Spring 2013 Assignment 3, v1.0 Due: Apr. 24th 11:59PM

CS 5540 Spring 2013 Assignment 3, v1.0 Due: Apr. 24th 11:59PM 1 Introduction In this programming project, we are going to do a simple image segmentation task. Given a grayscale image with a bright object against a dark background and we are going to do a binary decision

More information

The Rational Zero Theorem

The Rational Zero Theorem The Rational Zero Theorem Our goal in this section is to learn how we can ind the rational zeros o the polynomials. For example: x = x 4 + x x x + ( ) We could randomly try some actors and use synthetic

More information

Image Segmentation continued Graph Based Methods. Some slides: courtesy of O. Capms, Penn State, J.Ponce and D. Fortsyth, Computer Vision Book

Image Segmentation continued Graph Based Methods. Some slides: courtesy of O. Capms, Penn State, J.Ponce and D. Fortsyth, Computer Vision Book Image Segmentation continued Graph Based Methods Some slides: courtesy of O. Capms, Penn State, J.Ponce and D. Fortsyth, Computer Vision Book Previously Binary segmentation Segmentation by thresholding

More information

Supplementary Material: Adaptive and Move Making Auxiliary Cuts for Binary Pairwise Energies

Supplementary Material: Adaptive and Move Making Auxiliary Cuts for Binary Pairwise Energies Supplementary Material: Adaptive and Move Making Auxiliary Cuts for Binary Pairwise Energies Lena Gorelick Yuri Boykov Olga Veksler Computer Science Department University of Western Ontario Abstract Below

More information

Higher-order models in Computer Vision

Higher-order models in Computer Vision Chapter 1 Higher-order models in Computer Vision PUSHMEET KOHLI Machine Learning and Perception Microsoft Research Cambridge, UK Email: pkohli@microsoft.com CARSTEN ROTHER Machine Learning and Perception

More information

Srikumar Ramalingam. Review. 3D Reconstruction. Pose Estimation Revisited. School of Computing University of Utah

Srikumar Ramalingam. Review. 3D Reconstruction. Pose Estimation Revisited. School of Computing University of Utah School of Computing University of Utah Presentation Outline 1 2 3 Forward Projection (Reminder) u v 1 KR ( I t ) X m Y m Z m 1 Backward Projection (Reminder) Q K 1 q Q K 1 u v 1 What is pose estimation?

More information

Variable Grouping for Energy Minimization

Variable Grouping for Energy Minimization Variable Grouping for Energy Minimization Taesup Kim, Sebastian Nowozin, Pushmeet Kohli, Chang D. Yoo Dept. of Electrical Engineering, KAIST, Daejeon Korea Microsoft Research Cambridge, Cambridge UK kim.ts@kaist.ac.kr,

More information

Graph Cuts vs. Level Sets. part I Basics of Graph Cuts

Graph Cuts vs. Level Sets. part I Basics of Graph Cuts ECCV 2006 tutorial on Graph Cuts vs. Level Sets part I Basics of Graph Cuts Yuri Boykov University of Western Ontario Daniel Cremers University of Bonn Vladimir Kolmogorov University College London Graph

More information

Segmentation with non-linear constraints on appearance, complexity, and geometry

Segmentation with non-linear constraints on appearance, complexity, and geometry IPAM February 2013 Western Univesity Segmentation with non-linear constraints on appearance, complexity, and geometry Yuri Boykov Andrew Delong Lena Gorelick Hossam Isack Anton Osokin Frank Schmidt Olga

More information

Figure : Example Precedence Graph

Figure : Example Precedence Graph CS787: Advanced Algorithms Topic: Scheduling with Precedence Constraints Presenter(s): James Jolly, Pratima Kolan 17.5.1 Motivation 17.5.1.1 Objective Consider the problem of scheduling a collection of

More information

Combining Appearance and Structure from Motion Features for Road Scene Understanding

Combining Appearance and Structure from Motion Features for Road Scene Understanding STURGESS et al.: COMBINING APPEARANCE AND SFM FEATURES 1 Combining Appearance and Structure from Motion Features for Road Scene Understanding Paul Sturgess paul.sturgess@brookes.ac.uk Karteek Alahari karteek.alahari@brookes.ac.uk

More information

Curvature Prior for MRF-based Segmentation and Shape Inpainting

Curvature Prior for MRF-based Segmentation and Shape Inpainting Curvature Prior for MRF-based Segmentation and Shape Inpainting Alexander Shekhovtsov, Pushmeet Kohli +, Carsten Rother + Center for Machine Perception, Czech Technical University, Microsoft Research Cambridge,

More information

Geolocalization using Skylines from Omni-Images

Geolocalization using Skylines from Omni-Images Geolocalization using Skylines from Omni-Images Srikumar Ramalingam 1 Sofien Bouaziz 1&2 Peter Sturm 3 Matthew Brand 1 1 Mitsubishi Electric Research Lab (MERL), Cambridge, MA, USA 2 Virtual Reality Lab,

More information

CSE 417 Network Flows (pt 3) Modeling with Min Cuts

CSE 417 Network Flows (pt 3) Modeling with Min Cuts CSE 417 Network Flows (pt 3) Modeling with Min Cuts Reminders > HW6 is due on Friday start early bug fixed on line 33 of OptimalLineup.java: > change true to false Review of last two lectures > Defined

More information

Segmentation Based Stereo. Michael Bleyer LVA Stereo Vision

Segmentation Based Stereo. Michael Bleyer LVA Stereo Vision Segmentation Based Stereo Michael Bleyer LVA Stereo Vision What happened last time? Once again, we have looked at our energy function: E ( D) = m( p, dp) + p I < p, q > We have investigated the matching

More information

and 6.855J. The Successive Shortest Path Algorithm and the Capacity Scaling Algorithm for the Minimum Cost Flow Problem

and 6.855J. The Successive Shortest Path Algorithm and the Capacity Scaling Algorithm for the Minimum Cost Flow Problem 15.082 and 6.855J The Successive Shortest Path Algorithm and the Capacity Scaling Algorithm for the Minimum Cost Flow Problem 1 Pseudo-Flows A pseudo-flow is a "flow" vector x such that 0 x u. Let e(i)

More information

Tutorial for Algorithm s Theory Problem Set 5. January 17, 2013

Tutorial for Algorithm s Theory Problem Set 5. January 17, 2013 Tutorial for Algorithm s Theory Problem Set 5 January 17, 2013 Exercise 1: Maximum Flow Algorithms Consider the following flow network: a) Solve the maximum flow problem on the above network by using the

More information

ADVANCED MACHINE LEARNING MACHINE LEARNING. Kernel for Clustering kernel K-Means

ADVANCED MACHINE LEARNING MACHINE LEARNING. Kernel for Clustering kernel K-Means 1 MACHINE LEARNING Kernel for Clustering ernel K-Means Outline of Today s Lecture 1. Review principle and steps of K-Means algorithm. Derive ernel version of K-means 3. Exercise: Discuss the geometrical

More information

Notes on Minimum Cuts and Modular Functions

Notes on Minimum Cuts and Modular Functions Notes on Minimum Cuts and Modular Functions 1 Introduction The following are my notes on Cunningham s paper [1]. Given a submodular function f and a set S, submodular minimisation is the problem of finding

More information

Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction

Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction Nikolay Savinov 1, L ubor Ladický 1, Christian Häne and Marc Pollefeys ETH Zürich, Switzerland nikolay.savinov,lubor.ladicky,christian.haene,marc.pollefeys}@inf.ethz.ch

More information

ES 240: Scientific and Engineering Computation. a function f(x) that can be written as a finite series of power functions like

ES 240: Scientific and Engineering Computation. a function f(x) that can be written as a finite series of power functions like Polynomial Deinition a unction () that can be written as a inite series o power unctions like n is a polynomial o order n n ( ) = A polynomial is represented by coeicient vector rom highest power. p=[3-5

More information

9.8 Graphing Rational Functions

9.8 Graphing Rational Functions 9. Graphing Rational Functions Lets begin with a deinition. Deinition: Rational Function A rational unction is a unction o the orm P where P and Q are polynomials. Q An eample o a simple rational unction

More information

Cosegmentation Revisited: Models and Optimization

Cosegmentation Revisited: Models and Optimization Cosegmentation Revisited: Models and Optimization Sara Vicente 1, Vladimir Kolmogorov 1, and Carsten Rother 2 1 University College London 2 Microsoft Research Cambridge Abstract. The problem of cosegmentation

More information

9.3 Transform Graphs of Linear Functions Use this blank page to compile the most important things you want to remember for cycle 9.

9.3 Transform Graphs of Linear Functions Use this blank page to compile the most important things you want to remember for cycle 9. 9. Transorm Graphs o Linear Functions Use this blank page to compile the most important things you want to remember or cycle 9.: Sec Math In-Sync by Jordan School District, Utah is licensed under a 6 Function

More information

CS242: Probabilistic Graphical Models Lecture 3: Factor Graphs & Variable Elimination

CS242: Probabilistic Graphical Models Lecture 3: Factor Graphs & Variable Elimination CS242: Probabilistic Graphical Models Lecture 3: Factor Graphs & Variable Elimination Instructor: Erik Sudderth Brown University Computer Science September 11, 2014 Some figures and materials courtesy

More information

Human detection using histogram of oriented gradients. Srikumar Ramalingam School of Computing University of Utah

Human detection using histogram of oriented gradients. Srikumar Ramalingam School of Computing University of Utah Human detection using histogram of oriented gradients Srikumar Ramalingam School of Computing University of Utah Reference Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection,

More information

Inference in Higher Order MRF-MAP Problems with Small and Large Cliques

Inference in Higher Order MRF-MAP Problems with Small and Large Cliques Inference in Higher Order MRF-MAP Problems with Small and Large Cliques Ishant Shanu 1 Chetan Arora 1 S.N. Maheshwari 2 1 IIIT Delhi 2 IIT Delhi Abstract Higher Order MRF-MAP formulation has been a popular

More information

Geometric and Semantic 3D Reconstruction: Part 4A: Volumetric Semantic 3D Reconstruction. CVPR 2017 Tutorial Christian Häne UC Berkeley

Geometric and Semantic 3D Reconstruction: Part 4A: Volumetric Semantic 3D Reconstruction. CVPR 2017 Tutorial Christian Häne UC Berkeley Geometric and Semantic 3D Reconstruction: Part 4A: Volumetric Semantic 3D Reconstruction CVPR 2017 Tutorial Christian Häne UC Berkeley Dense Multi-View Reconstruction Goal: 3D Model from Images (Depth

More information

Examples of P vs NP: More Problems

Examples of P vs NP: More Problems Examples of P vs NP: More Problems COMP1600 / COMP6260 Dirk Pattinson Australian National University Semester 2, 2017 Catch Up / Drop in Lab When Fridays, 15.00-17.00 Where N335, CSIT Building (bldg 108)

More information

Local Algorithms for Sparse Spanning Graphs

Local Algorithms for Sparse Spanning Graphs Local Algorithms for Sparse Spanning Graphs Reut Levi Dana Ron Ronitt Rubinfeld Intro slides based on a talk given by Reut Levi Minimum Spanning Graph (Spanning Tree) Local Access to a Minimum Spanning

More information

Prof. Feng Liu. Spring /17/2017. With slides by F. Durand, Y.Y. Chuang, R. Raskar, and C.

Prof. Feng Liu. Spring /17/2017. With slides by F. Durand, Y.Y. Chuang, R. Raskar, and C. Prof. Feng Liu Spring 2017 http://www.cs.pdx.edu/~fliu/courses/cs510/ 05/17/2017 With slides by F. Durand, Y.Y. Chuang, R. Raskar, and C. Rother Last Time Image segmentation Normalized cut and segmentation

More information

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning Topics Markov Random Fields: Inference Exact: VE Exact+Approximate: BP Readings: Barber 5 Dhruv Batra

More information

Project 2 due today Project 3 out today. Readings Szeliski, Chapter 10 (through 10.5)

Project 2 due today Project 3 out today. Readings Szeliski, Chapter 10 (through 10.5) Announcements Stereo Project 2 due today Project 3 out today Single image stereogram, by Niklas Een Readings Szeliski, Chapter 10 (through 10.5) Public Library, Stereoscopic Looking Room, Chicago, by Phillips,

More information

Object Stereo Joint Stereo Matching and Object Segmentation

Object Stereo Joint Stereo Matching and Object Segmentation Object Stereo Joint Stereo Matching and Object Segmentation Michael Bleyer 1 Carsten Rother 2 Pushmeet Kohli 2 Daniel Scharstein 3 Sudipta Sinha 4 1 Vienna University of Technology 2 Microsoft Research

More information

Larger K-maps. So far we have only discussed 2 and 3-variable K-maps. We can now create a 4-variable map in the

Larger K-maps. So far we have only discussed 2 and 3-variable K-maps. We can now create a 4-variable map in the EET 3 Chapter 3 7/3/2 PAGE - 23 Larger K-maps The -variable K-map So ar we have only discussed 2 and 3-variable K-maps. We can now create a -variable map in the same way that we created the 3-variable

More information

Extensions of submodularity and their application in computer vision

Extensions of submodularity and their application in computer vision Extensions of submodularity and their application in computer vision Vladimir Kolmogorov IST Austria Oxford, 20 January 2014 Linear Programming relaxation Popular approach: Basic LP relaxation (BLP) -

More information

MEDICAL IMAGE COMPUTING (CAP 5937) LECTURE 10: Medical Image Segmentation as an Energy Minimization Problem

MEDICAL IMAGE COMPUTING (CAP 5937) LECTURE 10: Medical Image Segmentation as an Energy Minimization Problem SPRING 06 MEDICAL IMAGE COMPUTING (CAP 97) LECTURE 0: Medical Image Segmentation as an Energy Minimization Problem Dr. Ulas Bagci HEC, Center for Research in Computer Vision (CRCV), University of Central

More information

Piecewise polynomial interpolation

Piecewise polynomial interpolation Chapter 2 Piecewise polynomial interpolation In ection.6., and in Lab, we learned that it is not a good idea to interpolate unctions by a highorder polynomials at equally spaced points. However, it transpires

More information

Primal Dual Schema Approach to the Labeling Problem with Applications to TSP

Primal Dual Schema Approach to the Labeling Problem with Applications to TSP 1 Primal Dual Schema Approach to the Labeling Problem with Applications to TSP Colin Brown, Simon Fraser University Instructor: Ramesh Krishnamurti The Metric Labeling Problem has many applications, especially

More information

intro, applications MRF, labeling... how it can be computed at all? Applications in segmentation: GraphCut, GrabCut, demos

intro, applications MRF, labeling... how it can be computed at all? Applications in segmentation: GraphCut, GrabCut, demos Image as Markov Random Field and Applications 1 Tomáš Svoboda, svoboda@cmp.felk.cvut.cz Czech Technical University in Prague, Center for Machine Perception http://cmp.felk.cvut.cz Talk Outline Last update:

More information

Local cues and global constraints in image understanding

Local cues and global constraints in image understanding Local cues and global constraints in image understanding Olga Barinova Lomonosov Moscow State University *Many slides adopted from the courses of Anton Konushin Image understanding «To see means to know

More information