1 / 65

Tractable Higher Order Models in Computer Vision ( Part II )

Tractable Higher Order Models in Computer Vision ( Part II ). Presented by Xiaodan Liang. Slides from Carsten Rother, Sebastian Nowozin , Pusohmeet Khli Microsoft Research Cambridge. Part II. Submodularity Move making algorithms Higher-order model : P n Potts model.

vine
Download Presentation

Tractable Higher Order Models in Computer Vision ( Part II )

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Tractable Higher Order Models in Computer Vision (Part II) Presented by Xiaodan Liang Slides from Carsten Rother,Sebastian Nowozin, PusohmeetKhli Microsoft Research Cambridge

  2. Part II • Submodularity • Move making algorithms • Higher-order model : Pn Potts model

  3. Feature selection

  4. Factoring distributions Problem inherently combinatorial!

  5. Example: Greedy algorithm for feature selection

  6. Key property: Diminishing returns Selection A = {} Selection B = {X2,X3} Y“Sick” Y“Sick” X2“Rash” X3“Male” X1“Fever” Adding X1will help a lot! Adding X1doesn’t help much Theorem [Krause, Guestrin UAI ‘05]: Information gain F(A) in Naïve Bayes models is submodular! New feature X1 + s B Large improvement Submodularity: A + s Small improvement

  7. ~63% Why is submodularity useful? Theorem [Nemhauser et al ‘78] Greedy maximization algorithm returns Agreedy: F(Agreedy) ¸ (1-1/e) max|A| k F(A) • Greedy algorithm gives near-optimal solution! • For info-gain: Guarantees best possible unless P = NP! [Krause, Guestrin UAI ’05]

  8. Submodularity in Machine Learning • Many ML problems are submodular, i.e., for F submodular require: • Minimization: A* = argmin F(A) • Structure learning (A* = argmin I(XA; XV\A)) • Clustering • MAP inference in Markov Random Fields • … • Maximization: A* = argmax F(A) • Feature selection • Active learning • Ranking • …

  9. Set functions

  10. A [ B AÅB Submodular set functions • Set function F on V is called submodular if • Equivalent diminishing returns characterization: + ¸ + B A + S B Large improvement Submodularity: A + S Small improvement

  11. Submodularity and supermodularity

  12. Example: Mutual information

  13. Closedness properties F1,…,Fm submodular functions on V and 1,…,m > 0 Then: F(A) = ii Fi(A) is submodular! Submodularity closed under nonnegative linear combinations! Extremely useful fact!! • F(A) submodular ) P() F(A) submodular! • Multicriterion optimization: F1,…,Fm submodular, i¸0 )i i Fi(A) submodular

  14. Submodularity and Concavity g(|A|) |A|

  15. Maximum of submodular functions Suppose F1(A) and F2(A) submodular. Is F(A) = max(F1(A),F2(A))submodular? F(A) = max(F1(A),F2(A)) F1(A) F2(A) |A| max(F1,F2) not submodular in general!

  16. Minimum of submodular functions Well, maybe F(A) = min(F1(A),F2(A)) instead? F({b}) – F(;)=0 < F({a,b}) – F({a})=1 min(F1,F2) not submodular in general! But stay tuned

  17. Submodularity and convexity

  18. x{b} 2 1 x{a} -1 0 1 -2 The submodular polyhedron PF Example: V = {a,b} x({b}) · F({b}) PF x({a,b}) · F({a,b}) x({a}) · F({a})

  19. Lovasz extension

  20. w{b} 2 1 w{a} -1 0 1 -2 Example: Lovasz extension g(w) = max {wT x: x2PF} g([0,1]) = [0,1]T [-2,2] = 2 = F({b}) g([1,1]) = [1,1]T [-1,1] = 0 = F({a,b}) [-2,2] {b} {a,b} [-1,1] w=[0,1]want g(w) {} {a} Greedy ordering:e1 = b, e2 = a  w(e1)=1 > w(e2)=0 xw(e1)=F({b})-F(;)=2 xw(e2)=F({b,a})-F({b})=-2  xw=[-2,2]

  21. Why is this useful? Theorem [Lovasz ’83]:g(w) attains its minimum in [0,1]n at a corner! If we can minimize g on [0,1]n, can minimize F…(at corners, g and F take same values) g(w) convex (and efficient to evaluate) F(A) submodular Does the converse also hold? No, consider g(w1,w2,w3) = max(w1,w2+w3) {a} {b} {c} F({a,b})-F({a})=0 < F({a,b,c})-F({a,c})=1

  22. Minimizing a submodular function Ellipsoid algorithm Interior Points algorithm

  23. Example: Image denoising

  24. Y1 Y2 Y3 X1 X2 X3 Y4 Y5 Y6 X4 X5 X6 Y7 Y8 Y9 X7 X8 X9 Example: Image denoising Pairwise Markov Random Field P(x1,…,xn,y1,…,yn) = i,ji,j(yi,yj) ii(xi,yi) Wantargmaxy P(y | x) =argmaxy log P(x,y) =argminyi,j Ei,j(yi,yj)+i Ei(yi) Ei,j(yi,yj) = -log i,j(yi,yj) Xi: noisy pixels Yi: “true” pixels When is this MAP inference efficiently solvable(in high treewidth graphical models)?

  25. MAP inference in Markov Random Fields[Kolmogorov et al, PAMI ’04, see also: Hammer, Ops Res ‘65]

  26. Constrained minimization

  27. Part II • Submodularity • Move making algorithms • Higher-order model : Pn Potts model

  28. Multi-Label problems

  29. Move making expansions move and swap move for this problem

  30. Metric and Semi metric Potential functions

  31. if the pairwise potential functions define a metric then the energy function in equation (8) can be approximately minimized using alpha expansions. • if pairwise potential functions defines a semi-metric, it can be minimized using alpha beta-swaps.

  32. Move Energy • Each move: • A transformation function: • The energy of a move t: • The optimal move: Submodular set functions play an important role in energy minimization as they can be minimized in polynomial time

  33. The swap move algorithm

  34. The expansion move algorithm

  35. Higher order potential • The class of higher order clique potentials for which the expansion and swap moves can be computed in polynomial time The clique potential take the form:

  36. Question you should be asking: • Show that move energy is submodularfor all xc Can my higher order potential be solved using α-expansions?

  37. Moves for Higher Order Potentials • Form of the Higher Order Potentials Clique Inconsistency function: Pairwise potential: xj xi xk Sum Form c xm xl Max Form

  38. Theoretical Results: Swap • Move energy is always submodular if non-decreasing concave. proofs

  39. Condition for Swap move Concave Function:

  40. Prove • all projections on two variables of any alpha beta-swap move energy are submodular. • The cost of any configuration

  41. substitute Constraints 1: Lema 1: Constraints2: The theorem is true

  42. Condition for alpha expansion • Metric:

  43. Moves for Higher Order Potentials • Form of the Higher Order Potentials Clique Inconsistency function: Pairwise potential: xj xi xk Sum Form c xm xl Max Form

  44. Part II • Submodularity • Move making algorithms • Higher-order model : Pn Potts model

  45. Image Segmentation n = number of pixels E(X) = ∑ ci xi + ∑dij|xi-xj| E: {0,1}n→R 0 →fg, 1→bg i i,j Image Segmentation Unary Cost [Boykov and Jolly ‘ 01] [Blake et al. ‘04] [Rotheret al.`04]

  46. Pn Potts Potentials Patch Dictionary (Tree) { 0 if xi = 0, i ϵ p Cmax otherwise h(Xp) = Cmax 0 p • [slide credits: Kohli]

  47. Pn Potts Potentials n = number of pixels E: {0,1}n→R 0 →fg, 1→bg E(X) = ∑ ci xi+ ∑dij|xi-xj| +∑hp(Xp) i i,j p { 0 if xi = 0, i ϵ p Cmax otherwise h(Xp) = p • [slide credits: Kohli]

  48. Theoretical Results: Expansion • Move energy is always submodular if increasing linear See paper for proofs

  49. PN Potts Model c

More Related