Improved Moves for Truncated Convex Models M. Pawan Kumar Philip Torr.

Slides:



Advertisements
Similar presentations
MAP Estimation Algorithms in M. Pawan Kumar, University of Oxford Pushmeet Kohli, Microsoft Research Computer Vision - Part I.
Advertisements

MAP Estimation Algorithms in
POSE–CUT Simultaneous Segmentation and 3D Pose Estimation of Humans using Dynamic Graph Cuts Mathieu Bray Pushmeet Kohli Philip H.S. Torr Department of.
Mean-Field Theory and Its Applications In Computer Vision1 1.
Algorithms for MAP estimation in Markov Random Fields Vladimir Kolmogorov University College London Tutorial at GDR (Optimisation Discrète, Graph Cuts.
O BJ C UT M. Pawan Kumar Philip Torr Andrew Zisserman UNIVERSITY OF OXFORD.
Solving Markov Random Fields using Second Order Cone Programming Relaxations M. Pawan Kumar Philip Torr Andrew Zisserman.
Solving Markov Random Fields using Dynamic Graph Cuts & Second Order Cone Programming Relaxations M. Pawan Kumar, Pushmeet Kohli Philip Torr.
ICCV 2007 tutorial Part III Message-passing algorithms for energy minimization Vladimir Kolmogorov University College London.
The University of Ontario CS 4487/9687 Algorithms for Image Analysis Multi-Label Image Analysis Problems.
An Analysis of Convex Relaxations (PART I) Minimizing Higher Order Energy Functions (PART 2) Philip Torr Work in collaboration with: Pushmeet Kohli, Srikumar.
C. Olsson Higher-order and/or non-submodular optimization: Yuri Boykov jointly with Western University Canada O. Veksler Andrew Delong L. Gorelick C. NieuwenhuisE.
1 s-t Graph Cuts for Binary Energy Minimization  Now that we have an energy function, the big question is how do we minimize it? n Exhaustive search is.
Probabilistic Inference Lecture 1
Learning with Inference for Discrete Graphical Models Nikos Komodakis Pawan Kumar Nikos Paragios Ramin Zabih (presenter)
1 Can this be generalized?  NP-hard for Potts model [K/BVZ 01]  Two main approaches 1. Exact solution [Ishikawa 03] Large graph, convex V (arbitrary.
Robust Higher Order Potentials For Enforcing Label Consistency
An Analysis of Convex Relaxations M. Pawan Kumar Vladimir Kolmogorov Philip Torr for MAP Estimation.
P 3 & Beyond Solving Energies with Higher Order Cliques Pushmeet Kohli Pawan Kumar Philip H. S. Torr Oxford Brookes University CVPR 2007.
2010/5/171 Overview of graph cuts. 2010/5/172 Outline Introduction S-t Graph cuts Extension to multi-label problems Compare simulated annealing and alpha-
Stereo & Iterative Graph-Cuts Alex Rav-Acha Vision Course Hebrew University.
Efficiently Solving Convex Relaxations M. Pawan Kumar University of Oxford for MAP Estimation Philip Torr Oxford Brookes University.
Graph Cut based Inference with Co-occurrence Statistics Ľubor Ladický, Chris Russell, Pushmeet Kohli, Philip Torr.
Stereo Computation using Iterative Graph-Cuts
What Energy Functions Can be Minimized Using Graph Cuts? Shai Bagon Advanced Topics in Computer Vision June 2010.
Relaxations and Moves for MAP Estimation in MRFs M. Pawan Kumar STANFORDSTANFORD Vladimir KolmogorovPhilip TorrDaphne Koller.
Hierarchical Graph Cuts for Semi-Metric Labeling M. Pawan Kumar Joint work with Daphne Koller.
Measuring Uncertainty in Graph Cut Solutions Pushmeet Kohli Philip H.S. Torr Department of Computing Oxford Brookes University.
Graph-Cut Algorithm with Application to Computer Vision Presented by Yongsub Lim Applied Algorithm Laboratory.
Computer vision: models, learning and inference
Extensions of submodularity and their application in computer vision
MAP Estimation Algorithms in M. Pawan Kumar, University of Oxford Pushmeet Kohli, Microsoft Research Computer Vision - Part I.
Multiplicative Bounds for Metric Labeling M. Pawan Kumar École Centrale Paris École des Ponts ParisTech INRIA Saclay, Île-de-France Joint work with Phil.
Probabilistic Inference Lecture 4 – Part 2 M. Pawan Kumar Slides available online
Fast Approximate Energy Minimization via Graph Cuts
A Selective Overview of Graph Cut Energy Minimization Algorithms Ramin Zabih Computer Science Department Cornell University Joint work with Yuri Boykov,
Mutual Information-based Stereo Matching Combined with SIFT Descriptor in Log-chromaticity Color Space Yong Seok Heo, Kyoung Mu Lee, and Sang Uk Lee.
City University of Hong Kong 18 th Intl. Conf. Pattern Recognition Self-Validated and Spatially Coherent Clustering with NS-MRF and Graph Cuts Wei Feng.
Multiplicative Bounds for Metric Labeling M. Pawan Kumar École Centrale Paris Joint work with Phil Torr, Daphne Koller.
Rounding-based Moves for Metric Labeling M. Pawan Kumar Center for Visual Computing Ecole Centrale Paris.
Lena Gorelick joint work with O. Veksler I. Ben Ayed A. Delong Y. Boykov.
Discrete Optimization Lecture 2 – Part I M. Pawan Kumar Slides available online
Probabilistic Inference Lecture 3 M. Pawan Kumar Slides available online
Algorithms for MAP estimation in Markov Random Fields Vladimir Kolmogorov University College London.
Discrete Optimization in Computer Vision M. Pawan Kumar Slides will be available online
Discrete Optimization Lecture 3 – Part 1 M. Pawan Kumar Slides available online
1 Markov Random Fields with Efficient Approximations Yuri Boykov, Olga Veksler, Ramin Zabih Computer Science Department CORNELL UNIVERSITY.
Fast and accurate energy minimization for static or time-varying Markov Random Fields (MRFs) Nikos Komodakis (Ecole Centrale Paris) Nikos Paragios (Ecole.
Probabilistic Inference Lecture 5 M. Pawan Kumar Slides available online
Efficient Discriminative Learning of Parts-based Models M. Pawan Kumar Andrew Zisserman Philip Torr
O BJ C UT M. Pawan Kumar Philip Torr Andrew Zisserman UNIVERSITY OF OXFORD.
Tractable Higher Order Models in Computer Vision (Part II) Slides from Carsten Rother, Sebastian Nowozin, Pusohmeet Khli Microsoft Research Cambridge Presented.
The University of Ontario CS 4487/9687 Algorithms for Image Analysis Multi-Label Image Analysis Problems.
Discrete Optimization Lecture 2 – Part 2 M. Pawan Kumar Slides available online
Inference for Learning Belief Propagation. So far... Exact methods for submodular energies Approximations for non-submodular energies Move-making ( N_Variables.
Probabilistic Inference Lecture 2 M. Pawan Kumar Slides available online
Discrete Optimization Lecture 1 M. Pawan Kumar Slides available online
Pushmeet Kohli. E(X) E: {0,1} n → R 0 → fg 1 → bg Image (D) n = number of pixels [Boykov and Jolly ‘ 01] [Blake et al. ‘04] [Rother, Kolmogorov and.
A global approach Finding correspondence between a pair of epipolar lines for all pixels simultaneously Local method: no guarantee we will have one to.
Graph Algorithms for Vision Amy Gale November 5, 2002.
Tightening LP Relaxations for MAP using Message-Passing David Sontag Joint work with Talya Meltzer, Amir Globerson, Tommi Jaakkola, and Yair Weiss.
Linear Solution to Scale and Rotation Invariant Object Matching Hao Jiang and Stella X. Yu Computer Science Department Boston College.
MAP Estimation of Semi-Metric MRFs via Hierarchical Graph Cuts M. Pawan Kumar Daphne Koller Aim: To obtain accurate, efficient maximum a posteriori (MAP)
Rounding-based Moves for Metric Labeling M. Pawan Kumar École Centrale Paris INRIA Saclay, Île-de-France.
Energy minimization Another global approach to improve quality of correspondences Assumption: disparities vary (mostly) smoothly Minimize energy function:
Alexander Shekhovtsov and Václav Hlaváč
Markov Random Fields with Efficient Approximations
Efficient Graph Cut Optimization for Full CRFs with Quantized Edges
Discrete Optimization Methods Basic overview of graph cuts
MAP Estimation of Semi-Metric MRFs via Hierarchical Graph Cuts
Presentation transcript:

Improved Moves for Truncated Convex Models M. Pawan Kumar Philip Torr

Aim Efficient, accurate MAP for truncated convex models V1V1 V2V2 ……… …………… …………… …………VnVn Random Variables V = { V 1, V 2, …, V n } Edges E define neighbourhood

Aim VaVa VbVb lili lklk  ab;ik Accurate, efficient MAP for truncated convex models  ab;ik = w ab min{ d(i-k), M }  ab;ik i-k w ab is non-negative Truncated Linear i-k  ab;ik Truncated Quadratic d(.) is convex  a;i  b;k

Motivation Low-level Vision Smoothly varying regions Sharp edges between regions min{ |i-k|, M} Boykov, Veksler & Zabih 1998 Well-researched !!

Things We Know NP-hard problem - Can only get approximation Best possible integrality gap - LP relaxation Manokaran et al., 2008 Solve using TRW-S, DD, PP Slower than graph-cuts Use Range Move - Veksler, 2007 None of the guarantees of LP

Real Motivation Gaps in Move-Making Literature LP Move- Making Potts Truncated Linear Truncated Quadratic 2 Multiplicative Bounds 2 + √2 O(√M) Chekuri et al., 2001

Real Motivation Gaps in Move-Making Literature LP Move- Making Potts Truncated Linear Truncated Quadratic 2 Multiplicative Bounds √2 2M O(√M)- Boykov, Veksler and Zabih, 1999

Real Motivation Gaps in Move-Making Literature LP Move- Making Potts Truncated Linear Truncated Quadratic 2 Multiplicative Bounds √2 4 O(√M)- Gupta and Tardos, 2000

Real Motivation Gaps in Move-Making Literature LP Move- Making Potts Truncated Linear Truncated Quadratic 2 Multiplicative Bounds √2 4 O(√M)2M Komodakis and Tziritas, 2005

Real Motivation Gaps in Move-Making Literature LP Move- Making Potts Truncated Linear Truncated Quadratic 2 Multiplicative Bounds √2 O(√M) 2 + √2 O(√M)

Outline Move Space Graph Construction Sketch of the Analysis Results

Move Space VaVa VbVb Initialize the labelling Choose interval I of L’ labels Each variable can Retain old label Choose a label from I Choose best labelling Iterate over intervals

Outline Move Space Graph Construction Sketch of the Analysis Results

Two Problems VaVa VbVb Choose interval I of L’ labels Each variable can Retain old label Choose a label from I Choose best labelling Large L’ => Non-submodular Non-submodular

First Problem VaVa VbVb Submodular problem Ishikawa, 2003; Veksler, 2007

First Problem VaVa VbVb Non-submodular Problem

First Problem VaVa VbVb Submodular problem Veksler, 2007

First Problem VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb Model unary potentials exactly a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb Similarly for V b a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb Model convex pairwise costs a m+1 a m+2 anan t b m+1 b m+2 bnbn

First Problem VaVa VbVb Overestimated pairwise potentials Wanted to model  ab;ik = w ab min{ d(i-k), M } For all l i, l k  I Have modelled  ab;ik = w ab d(i-k) For all l i, l k  I

Second Problem VaVa VbVb Choose interval I of L’ labels Each variable can Retain old label Choose a label from I Choose best labelling Non-submodular problem !!

Second Problem - Case 1 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn s ∞∞ Both previous labels lie in interval

Second Problem - Case 1 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn s ∞∞ w ab d(i-k)

Second Problem - Case 2 VaVa VbVb Only previous label of V a lies in interval a m+1 a m+2 anan t b m+1 b m+2 bnbn s ∞ ubub

Second Problem - Case 2 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn u b : unary potential of previous label of V b M s ∞ ubub

Second Problem - Case 2 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn M w ab d(i-k) s ∞ ubub

Second Problem - Case 2 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn M w ab ( d(i-m-1) + M ) s ∞ ubub

Second Problem - Case 3 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn Only previous label of V b lies in interval

Second Problem - Case 3 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn s uaua ∞ u a : unary potential of previous label of V a M

Second Problem - Case 4 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn Both previous labels do not lie in interval

Second Problem - Case 4 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn s uaua ubub P ab : pairwise potential for previous labels ab P ab M M

Second Problem - Case 4 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn w ab d(i-k) s uaua ubub ab P ab M M

Second Problem - Case 4 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn w ab ( d(i-m-1) + M ) s uaua ubub ab P ab M M

Second Problem - Case 4 VaVa VbVb a m+1 a m+2 anan t b m+1 b m+2 bnbn P ab s uaua ubub ab P ab M M

Graph Construction VaVa VbVb Find st-MINCUT. Retain old labelling if energy increases. a m+1 a m+2 anan b m+1 b m+2 bnbn t ITERATE

Outline Move Space Graph Construction Sketch of the Analysis Results

Analysis VaVa VbVb Current labelling f(.) Q C ≤ Q’ C VaVa VbVb Global Optimum f*(.) QPQP Previous labelling f’(.) VaVa VbVb

Analysis VaVa VbVb Current labelling f(.) Q C ≤ Q’ C VaVa VbVb Partially Optimal f’’(.) Previous labelling f’(.) VaVa VbVb Q’ 0 ≤

Analysis VaVa VbVb Current labelling f(.) Q P - Q’ C VaVa VbVb Partially Optimal f’’(.) Previous labelling f’(.) VaVa VbVb Q P - Q’ 0 ≥

Analysis VaVa VbVb Current labelling f(.) Q P - Q’ C VaVa VbVb Partially Optimal f’’(.) Local Optimal f’(.) VaVa VbVb Q P - Q’ 0 ≤ 0

Analysis VaVa VbVb Current labelling f(.) VaVa VbVb Partially Optimal f’’(.) Local Optimal f’(.) VaVa VbVb Q P - Q’ 0 ≤ 0 Take expectation over all intervals

Analysis Truncated Linear Q P ≤ 2 + max 2M, L’ L’M Q* L’ = M4Gupta and Tardos, 2000 L’ = √2M 2 + √2 Truncated Quadratic Q P ≤ O(√M) Q* L’ = √M

Outline Move Space Graph Construction Sketch of the Analysis Results

Synthetic Data - Truncated Linear Faster than TRW-S Comparable to Range Moves With LP Relaxation guarantees Time (sec) Energy

Synthetic Data - Truncated Quadratic Faster than TRW-S Comparable to Range Moves With LP Relaxation guarantees Time (sec) Energy

Stereo Correspondence Disparity Map Unary Potential: Similarity of pixel colour Pairwise Potential: Truncated convex

Stereo Correspondence AlgoEnergy1Time1Energy2Time2 Swap Exp TRW-S BP Range Our Teddy

Stereo Correspondence AlgoEnergy1Time1Energy2Time2 Swap Exp TRW-S BP Range Our Teddy

Stereo Correspondence AlgoEnergy1Time1Energy2Time2 Swap Exp TRW-S BP Range Our Tsukuba

Summary Moves that give LP guarantees Similar results to TRW-S Faster than TRW-S because of graph cuts

Questions Not Yet Answered Move-making gives LP guarantees –True for all MAP estimation problems? Huber function? Parallel Imaging Problem? Primal-dual method? Solving more complex relaxations?

Questions? Improved Moves for Truncated Convex Models Kumar and Torr, NIPS