SlideShare une entreprise Scribd logo
1  sur  115
Télécharger pour lire hors ligne
Remeshing
Variational methods
Bruno Lévy
ALICE Géométrie & Lumière
CENTRE INRIA Nancy Grand-Est
OVERVIEW
Part. 1. Introduction - Motivations
Part. 2. Blowing Bubbles
Part. 3. Least Action (the power of power diagrams)
Conclusions
Introduction - Motivations
1
1. Motivations
[Martinez, Dumas, Lefebvre]
1. Motivations
1. Motivations
1. Motivations
1. Motivations
René Descartes
Wonder
Shapes
Structures
Symmetry
Blowing Bubbles
Centroidal Voronoi Tesselations
2
Voronoi diagram: Vor(xi) = { x | d2(x,xi) < d2(x,xj) }
Chap. 2. Centroidal Voronoi Tesselation
Chap. 2.
Centroidal
Voronoi
Tessellation
Chap. 2. Centroidal Voronoi Tesselation
The classical method:
Lloyd algorithm = gradient descent
Chap. 2. Centroidal Voronoi Tesselation
The classical method:
Lloyd algorithm = gradient descent
Which objective function ?
Chap. 2. Centroidal Voronoi Tesselation
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
F=
∫Vor(i)
2
dxxi - x
i
Chap. 2. Centroidal Voronoi Tesselation
F=
∫Vor(i)
2
dxxi - x
i
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Chap. 2. Centroidal Voronoi Tesselation
F=
∫Vor(i)
2
dxxi - x
i
=
∫Inf i
2
dxxi - x
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Chap. 2. Centroidal Voronoi Tesselation
F=
∫Vor(i)
2
dxxi - x
i
=
∫Inf i
2
dxxi - x
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Chap. 2. Centroidal Voronoi Tesselation
F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al]
F=
∫Vor(i)
2
dxxi - x
i
=
∫Inf i
2
dxxi - x
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Chap. 2. Centroidal Voronoi Tesselation
F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al]
F=
∫Vor(i)
2
dxxi - x
i
=
∫Inf i
2
dxxi - x
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Volume of Vor(i)
Chap. 2. Centroidal Voronoi Tesselation
F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al]
F=
∫Vor(i)
2
dxxi - x
i
=
∫Inf i
2
dxxi - x
The classical method:
Lloyd algorithm = gradient descent
Which objective function ? This one (quantization noise power)
Why Lloyd works ? What is the gradient of F ?
Volume of Vor(i) Centroid of Vor(i)
Chap. 2. Centroidal Voronoi Tesselation
The classical method:
Lloyd algorithm = gradient descent
F=
∫Vor(i)
2
dxxi - x
i
F is smooth (C2) [Liu, Wang, L, Sun, Yan, Lu, Yang 2009]
Chap. 2. Centroidal Voronoi Tesselation
The classical method:
Lloyd algorithm = gradient descent
F=
∫Vor(i)
2
dxxi - x
i
F is smooth (C2) [Liu, Wang, L, Sun, Yan, Lu, Yang 2009]
[Yan, L, Liu, Sun and Wang 2009]
Chap. 2. Centroidal Voronoi Tesselation
F=
∫Vor(i)
2
dxxi - x
i
Blowing square bubbles
Chap. 2. Centroidal Voronoi Tesselation
p=2 p=4 p=8 ….
F=
∫Vor(i)
2
dxxi - x
i p
Chap. 2.
Chap. 2. Centroidal Voronoi Tesselation
[L, Liu, 2010]
Chap. 2. Centroidal Voronoi Tesselation
[L, Liu, 2010] [Ray, Sokolov, L, 2015, 2016]
(completely different method:
direction fields + global parameterization)
Least Action
Power Diagrams and Optimal Transport
3
Yann Brenier
The polar factorization theorem
(Brenier Transport)
Each time the Laplace operator is used in a PDE, it can be replaced with the
Monte-Ampère operator, and then interesting things occur
Euler
Lagrange
The Least Action Principle
Hamilton,
Legendre,
Maupertuis
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
Short summary of the 1st chapter of Landau,Lifshitz Course of Theoretical Physics
Euler
Lagrange
The Least Action Principle
Hamilton,
Legendre,
Maupertuis
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
position
Euler
Lagrange
The Least Action Principle
Hamilton,
Legendre,
Maupertuis
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
position
speed
Euler
Lagrange
The Least Action Principle
Hamilton,
Legendre,
Maupertuis
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
position
speed
time
Euler
Lagrange
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Hamilton,
Legendre,
Maupertuis
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
Axiom 2: The movement (time
evolution) of the physical system
minimizes the following integral
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
Axiom 2: The movement (time
evolution) of the physical system
minimizes the following integral
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists a function L(x,x,t) that describes the state
of a physical system
Axiom 2: The movement (time
evolution) of the physical system
minimizes the following integral
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
t=0
t=1
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
Isotropy of space =>
Preservation of angular momentum
∫t1
t2
L(x,x,t) dt
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
Isotropy of space =>
Preservation of angular momentum
Preserved quantities
Integrals of Motion
Noeter theorem
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
Isotropy of space =>
Preservation of angular momentum
Free particle:
Theorem 3: v = cte (Newton law I)
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
Isotropy of space =>
Preservation of angular momentum
Free particle:
Theorem 3: v = cte (Newton law I)
Expression of the Lagrangian:
L = ½ m v2
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Theorem 2:
∂L
∂x
x - L = cte
Homogeneity of time =>
Preservation of energy
Homogeneity of space =>
Preservation of momentum
Isotropy of space =>
Preservation of angular momentum
Free particle:
Theorem 3: v = cte (Newton law I)
Expression of the Lagrangian:
L = ½ m v2
Expression of the Energy:
E = ½ m v2
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Particle in a field:
Expression of the Lagrangian:
L = ½ m v2 – U(x)
Free particle:
Theorem 3: v = cte (Newton law I)
Expression of the Lagrangian:
L = ½ m v2
Expression of the Energy:
E = ½ m v2
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Particle in a field:
Expression of the Lagrangian:
L = ½ m v2 – U(x)
Expression of the Energy:
E = ½ m v2 + U(x)
Free particle:
Theorem 3: v = cte (Newton law I)
Expression of the Lagrangian:
L = ½ m v2
Expression of the Energy:
E = ½ m v2
The Least Action Principle
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. change of
Gallileo frame + hom. + isotrop. :
x’
t’ =
x+vt
t
Particle in a field:
Expression of the Lagrangian:
L = ½ m v2 – U(x)
Expression of the Energy:
E = ½ m v2 + U(x)
Theorem 4:
mx = - U (Newton law II)
Free particle:
Theorem 3: v = cte (Newton law I)
Expression of the Lagrangian:
L = ½ m v2
Expression of the Energy:
E = ½ m v2
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. Lorentz change of
frame
x’
t’ =
(x-vt) x γ
(t – vx/c2) x γ
γ = 1 / ( 1 – v2 / c2)
The Least Action Principle
(relativistic setting, just for fun)
√
The Least Action Principle
(relativistic setting, just for fun)
Axiom 1: There exists L
Axiom 2: The movement minimizes ∫ L
Theorem 1: (Lagrange equation):
∂L
∂x
d
dt
∂L
∂x
=
Axiom 3:
Invariance w.r.t. Lorentz change of
frame
x’
t’ =
(x-vt) x γ
(t – vx/c2) x γ
Theorem 5:
E = ½ γ m v2 + mc2
γ = 1 / ( 1 – v2 / c2)√
The Least Action Principle
(quantum physics setting, just for fun)
Fluids
FluidsLagrange point of view
FluidsLagrange point of view
ρ nb particles per square
Euler point of view
FluidsLagrange point of view
ρ(x,y,t) nb particles per square
Euler point of view
v(x,y,t) speed of the particle under
grid point (x,y) at time t
Fluids
ρ(x,y,t) nb particles per square
Euler point of view
v(x,y,t) speed of the particle under
grid point (x,y) at time t
Q1: how to compute the
acceleration of the particles
from v(x,y,t) ?
dv
dt
=
∂v v. v
∂t
+
Q2: incompressible fluids ?
div(v) = 0
Q3: mass preservation ?
d ρ
dt = - div(ρv)
(Continuity equation)
Fluids
Start with Lagrange coordinates:
particle trajectories: X(t,x)
Minimize
Action:
∫t1
t2
1/2
∫Ω
∂X (t,x)
∂t
dxdt
(ρ = cte)
s.t. X satisfies mass preservation
(X is measure-preserving, more on
this later)
2
Fluids
Start with Lagrange coordinates:
particle trajectories: X(t,x)
Minimize
Action:
∫t1
t2
1/2
∫Ω
∂X (t,x)
∂t
dxdt
(ρ = cte)
s.t. X satisfies mass preservation
(X is measure-preserving, more on
this later)
Acceleration of the
particle under the grid
2
Fluids
Start with Lagrange coordinates:
particle trajectories: X(t,x)
Minimize
Action:
∫t1
t2
1/2
∫Ω
∂X (t,x)
∂t
dxdt
(ρ = cte)
s.t. X satisfies mass preservation
(X is measure-preserving, more on
this later)
The Lagrange multiplier
for the constraint = pressure
2
?
T
Fluids – Benamou Brenier
ρ1 ρ2
?
T
Fluids – Benamou Brenier
Minimize
A(ρ,v) =
∫t1
t2
(t2-t1)
∫Ω
ρ(x,t) ||v(t,x)||2
dxdt
s.t. ρ(t1,.) = ρ1 ; ρ(t2,.) = ρ2 ; d ρ
dt
= - div(ρv)
ρ1 ρ2
?
T
Fluids – Benamou Brenier
Minimize
A(ρ,v) =
∫t1
t2
(t2-t1)
∫Ω
ρ(x,t) ||v(t,x)||2
dxdt
s.t. ρ(t1,.) = ρ1 ; ρ(t2,.) = ρ2 ; d ρ
dt
= - div(ρv)
ρ1 ρ2
Minimize C(T) =
∫Ω
|| x – T(x) ||2 dx
s.t. T is measure-preserving
ρ1(x)
Part. 3 Optimal Transport – Monge problem
A map T is a transport map between μ and ν if
μ(T-1(B)) = ν(B) for any Borel subset B of Y
(X;μ) (Y;ν)
Part. 3 Optimal Transport – Monge problem
A map T is a transport map between μ and ν if
μ(T-1(B)) = ν(B) for any Borel subset B
B
(X;μ) (Y;ν)
Part. 3 Optimal Transport – Monge problem
A map T is a transport map between μ and ν if
μ(T-1(B)) = ν(B) for any Borel subset B
B
T-1(B)
(X;μ) (Y;ν)
Part. 3 Optimal Transport – Monge problem
A map T is a transport map between μ and ν if
μ(T-1(B)) = ν(B) for any Borel subset B
(or ν = T#μ the pushforward of μ)
(X;μ) (Y;ν)
Part. 3 Optimal Transport – Monge problem
Monge problem (1787):
Find a transport map T that minimizes C(T) = ∫X || x – T(x) ||2 dμ(x)
(X;μ) (Y;ν)
Part. 3 Optimal Transport – Kantorovich
Monge problem:
Find a transport map T that minimizes C(T) = ∫X || x – T(x) ||2 dμ(x)
Kantorovich problem (1942):
Find a measure γ defined on X x Y
such that ∫x in X dγ(x,y) = dν(y)
and ∫y in Y dγ(x,y) = dμ(x)
that minimizes ∫∫X x Y || x – y ||2 dγ(x,y)
Part. 3 Optimal Transport – Kantorovich
Transport plan – example 1/2 : translation of a segment
Part. 3 Optimal Transport – Kantorovich
Transport plan – example 1/2 : translation of a segment
Part. 3 Optimal Transport – Kantorovich
Part. 3 Optimal Transport – Dual of Kantorovich
Dual formulation of Kantorovich problem:
Find a c-concave function ψ
that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν
ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform)
Part. 3 Optimal Transport – Dual of Kantorovich
Dual formulation of Kantorovich problem:
Find a c-concave function ψ
that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν
ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform)
What about our initial problem ?
T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) )
{{grad ψ(x) with ψ(x) := (½ x2- ψ(x))
[Brenier]
Part. 3 Optimal Transport – Dual of Kantorovich
Dual formulation of Kantorovich problem:
Find a c-concave function ψ
that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν
ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform)
What about our initial problem ?
T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) )
{{grad ψ(x) with ψ(x) := (½ x2- ψ(x))
When μ and ν have a density u and v, (H ψ(x)). v(grad ψ(x)) = u(x)
Monge-Ampère
equation
for all borel set A, ∫A dμ = ∫T(A) (|JT|) dν = ∫T(A) (H ψ ) dν
[Brenier]
Part. 3 Optimal Transport – Dual of Kantorovich
Dual formulation of Kantorovich problem:
Find a c-concave function ψ
that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν
ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform)
What about our initial problem ?
T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) )
{{grad ψ(x) with ψ(x) := (½ x2- ψ(x))
When μ and ν have a density u and v, (H ψ(x)). v(grad ψ(x)) = u(x)
Monge-Ampère
equation
for all borel set A, ∫A dμ = ∫T(A) (|JT|) dν = ∫T(A) (H ψ ) dν
[Brenier]
Part. 3 Optimal Transport – semi-discrete
∫X ψc (x)dμ + ∫Y ψ(y)dν
Sup
ψ Є ψc
(DMK)
(X;μ) (Y;ν)
Part. 3 Optimal Transport – semi-discrete
(X;μ) (Y;ν)
∑j ψ(yj) vj
∫X ψc (x)dμ + ∫Y ψ(y)dν
Sup
ψ Є ψc
(DMK)
Part. 3 Optimal Transport – semi-discrete
∫X ψc (x)dμ + ∫Y ψ(y)dν
Sup
ψ Є ψc
(DMK)
∑j ψ(yj) vj
Part. 3 Optimal Transport – semi-discrete
∫X ψc (x)dμ + ∫Y ψ(y)dν
Sup
ψ Є ψc
(DMK)
∫X inf yj ЄY [ || x – yj ||2 - ψ(yj) ] dμ
∑j ψ(yj) vj
Part. 3 Optimal Transport – semi-discrete
∫X ψc (x)dμ + ∫Y ψ(y)dν
Sup
ψ Є ψc
(DMK)
∑j ∫Lagψ(yj) || x – yj ||2 - ψ(yj) dμ
∫X inf yj ЄY [ || x – yj ||2 - ψ(yj) ] dμ
∑j ψ(yj) vj
Power diagram: Pow(xi) = { x | d2(x,xi) – ψi < d2(x,xj) – ψj }
Voronoi diagram: Vor(xi) = { x | d2(x,xi) < d2(x,xj) }
Part. 3 Power Diagrams
Part. 3 Power Diagrams
Part. 3 Optimal Transport
Theorem: (direct consequence of MK duality
alternative proof in [Aurenhammer, Hoffmann, Aronov 98] ):
Given a measure μ with density, a set of points (yj), a set of positive coefficients vj
such that ∑ vj =∫ dμ(x), it is possible to find the weights W = [ψ(y1) ψ(y2) … ψ(ym)]
such that the map TS
W is the unique optimal transport map
between μ and ν = ∑ vj δ(yj)
Part. 3 Optimal Transport
Optimal transport Centroidal Voronoi Tesselation
[L - A numerical Algorithm for L2 semi-discrete OT (ESAIM M2AN 2015)]
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
hi
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams & Transport
Part. 3 Power Diagrams and Transport
Part. 3 Optimal Transport – 2D examples
Numerical Experiment: A disk becomes two disks
Part. 3 Optimal Transport – 3D examples
Numerical Experiment: A sphere becomes two spheres
Part. 3 Optimal Transport – 3D examples
Numerical Experiment: Armadillo to sphere
Part. 3 Optimal Transport – 3D examples
Numerical Experiment: Other examples
Part. 3 Optimal Transport – 3D examples
Numerical Experiment: Varying density
Part. 1 Optimal Transport
[Schwartzburg et.al, SIGGRAPH 2014]
Part. 1 Optimal Transport
Optimal transport E.U.R.
let there be light !
Part. 1 Optimal Transport
Optimal transport E.U.R.
let there be light !
The millenium simulation project,
Max Planck Institute fur Astrophysik
pc/h : parsec (= 3.2 light years)
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
2015:
3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
2015:
3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
2015:
3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D:
1 million Dirac masses in 240 seconds
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
2015:
3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D:
1 million Dirac masses in 240 seconds
10 million Dirac masses in 90 minutes
Part. 4 Optimal Transport – 3D examples
Numerical Experiment: Performances
2002: several hours of supercomputer time were needed
for computing OT with a few thousand Dirac masses, with a combinatorial
algorithm in O(n2log(n))
2015:
3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D:
1 million Dirac masses in 240 seconds
10 million Dirac masses in 90 minutes
Semi-discrete OT is now scalable ! (new tool in Num. Ana. Toolbox)
Other topics
•Euler equation in more complicated setting:
[Merigot & Mirebeau, Merigot & Galouet]
•Using semi-discrete OT to solve other PDEs
[Benamou, Carlier, Merigot , Oudet]
•Faster solvers
Online resources
Source code: alice.loria.fr/software/geogram
Demos: www.loria.fr/~levy/GEOGRAM/
www.loria.fr/~levy/GLUP/
L., A numerical algorithm for semi-discrete L2 OT in 3D,
ESAIM Math. Modeling and Analysis, 2015
Video (presentation with the proofs)
www.loria.fr/~levy (Ecole d été calcul des variations 2015)
New projects:
MAGA Q. Mérigot - ANR
EXPLORAGRAM (Inria) with E. Schwindt, Q. Mérigot and J.-D. Benamou
merci
http://alice.loria.fr
Voronoy Story

Contenu connexe

Tendances

Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...
Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...
Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...Alessandro Palmeri
 
Phase Retrieval: Motivation and Techniques
Phase Retrieval: Motivation and TechniquesPhase Retrieval: Motivation and Techniques
Phase Retrieval: Motivation and TechniquesVaibhav Dixit
 
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...Transverse magnetic plane-wave scattering equations for infinite and semi-inf...
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...Yong Heui Cho
 
Accelerating Dynamic Time Warping Subsequence Search with GPU
Accelerating Dynamic Time Warping Subsequence Search with GPUAccelerating Dynamic Time Warping Subsequence Search with GPU
Accelerating Dynamic Time Warping Subsequence Search with GPUDavide Nardone
 
Fpw chapter 4 - digital ctrl of dynamic systems
Fpw chapter 4 - digital ctrl of dynamic systemsFpw chapter 4 - digital ctrl of dynamic systems
Fpw chapter 4 - digital ctrl of dynamic systemsAleksandar Micic
 
Lecture 15 DCT, Walsh and Hadamard Transform
Lecture 15 DCT, Walsh and Hadamard TransformLecture 15 DCT, Walsh and Hadamard Transform
Lecture 15 DCT, Walsh and Hadamard TransformVARUN KUMAR
 
No Cloning Theorem with essential Mathematics and Physics
No Cloning Theorem with essential Mathematics and PhysicsNo Cloning Theorem with essential Mathematics and Physics
No Cloning Theorem with essential Mathematics and PhysicsRitajit Majumdar
 
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...Yuki Oyama
 
Lossless image compression using new biorthogonal wavelets
Lossless image compression using new biorthogonal waveletsLossless image compression using new biorthogonal wavelets
Lossless image compression using new biorthogonal waveletssipij
 
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...TELKOMNIKA JOURNAL
 
13 fourierfiltrationen
13 fourierfiltrationen13 fourierfiltrationen
13 fourierfiltrationenhoailinhtinh
 
Elements Of Stochastic Processes
Elements Of Stochastic ProcessesElements Of Stochastic Processes
Elements Of Stochastic ProcessesMALAKI12003
 
IRJET - Some Results on Fuzzy Semi-Super Modular Lattices
IRJET - Some Results on Fuzzy Semi-Super Modular LatticesIRJET - Some Results on Fuzzy Semi-Super Modular Lattices
IRJET - Some Results on Fuzzy Semi-Super Modular LatticesIRJET Journal
 
Automated seismic-to-well ties?
Automated seismic-to-well ties?Automated seismic-to-well ties?
Automated seismic-to-well ties?UT Technology
 
EH1 - Reduced-order modelling for vibration energy harvesting
EH1 - Reduced-order modelling for vibration energy harvestingEH1 - Reduced-order modelling for vibration energy harvesting
EH1 - Reduced-order modelling for vibration energy harvestingUniversity of Glasgow
 

Tendances (20)

03 image transform
03 image transform03 image transform
03 image transform
 
Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...
Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...
Toward an Improved Computational Strategy for Vibration-Proof Structures Equi...
 
Phase Retrieval: Motivation and Techniques
Phase Retrieval: Motivation and TechniquesPhase Retrieval: Motivation and Techniques
Phase Retrieval: Motivation and Techniques
 
report
reportreport
report
 
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...Transverse magnetic plane-wave scattering equations for infinite and semi-inf...
Transverse magnetic plane-wave scattering equations for infinite and semi-inf...
 
Accelerating Dynamic Time Warping Subsequence Search with GPU
Accelerating Dynamic Time Warping Subsequence Search with GPUAccelerating Dynamic Time Warping Subsequence Search with GPU
Accelerating Dynamic Time Warping Subsequence Search with GPU
 
Fpw chapter 4 - digital ctrl of dynamic systems
Fpw chapter 4 - digital ctrl of dynamic systemsFpw chapter 4 - digital ctrl of dynamic systems
Fpw chapter 4 - digital ctrl of dynamic systems
 
Lecture 15 DCT, Walsh and Hadamard Transform
Lecture 15 DCT, Walsh and Hadamard TransformLecture 15 DCT, Walsh and Hadamard Transform
Lecture 15 DCT, Walsh and Hadamard Transform
 
No Cloning Theorem with essential Mathematics and Physics
No Cloning Theorem with essential Mathematics and PhysicsNo Cloning Theorem with essential Mathematics and Physics
No Cloning Theorem with essential Mathematics and Physics
 
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...
Yuki Oyama - Incorporating context-dependent energy into the pedestrian dynam...
 
Lossless image compression using new biorthogonal wavelets
Lossless image compression using new biorthogonal waveletsLossless image compression using new biorthogonal wavelets
Lossless image compression using new biorthogonal wavelets
 
MM2020-AV
MM2020-AVMM2020-AV
MM2020-AV
 
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...
Bayesian Segmentation in Signal with Multiplicative Noise Using Reversible Ju...
 
13 fourierfiltrationen
13 fourierfiltrationen13 fourierfiltrationen
13 fourierfiltrationen
 
Elements Of Stochastic Processes
Elements Of Stochastic ProcessesElements Of Stochastic Processes
Elements Of Stochastic Processes
 
IRJET - Some Results on Fuzzy Semi-Super Modular Lattices
IRJET - Some Results on Fuzzy Semi-Super Modular LatticesIRJET - Some Results on Fuzzy Semi-Super Modular Lattices
IRJET - Some Results on Fuzzy Semi-Super Modular Lattices
 
1 tracking systems1
1 tracking systems11 tracking systems1
1 tracking systems1
 
Automated seismic-to-well ties?
Automated seismic-to-well ties?Automated seismic-to-well ties?
Automated seismic-to-well ties?
 
Image transforms
Image transformsImage transforms
Image transforms
 
EH1 - Reduced-order modelling for vibration energy harvesting
EH1 - Reduced-order modelling for vibration energy harvestingEH1 - Reduced-order modelling for vibration energy harvesting
EH1 - Reduced-order modelling for vibration energy harvesting
 

Similaire à Voronoy Story

Calculus of variations
Calculus of variationsCalculus of variations
Calculus of variationsSolo Hermelin
 
Effective Field Theories in the Quest for BSM Physics
Effective Field Theories in the Quest for BSM PhysicsEffective Field Theories in the Quest for BSM Physics
Effective Field Theories in the Quest for BSM PhysicsRaquel Gomez Ambrosio
 
Slides econometrics-2018-graduate-4
Slides econometrics-2018-graduate-4Slides econometrics-2018-graduate-4
Slides econometrics-2018-graduate-4Arthur Charpentier
 
Admissions in india 2015
Admissions in india 2015Admissions in india 2015
Admissions in india 2015Edhole.com
 
Admission in india 2015
Admission in india 2015Admission in india 2015
Admission in india 2015Edhole.com
 
Causal Dynamical Triangulations
Causal Dynamical TriangulationsCausal Dynamical Triangulations
Causal Dynamical TriangulationsRene García
 
Limits and Continuity of Functions
Limits and Continuity of Functions Limits and Continuity of Functions
Limits and Continuity of Functions OlooPundit
 
Top schools in delhi ncr
Top schools in delhi ncrTop schools in delhi ncr
Top schools in delhi ncrEdhole.com
 
Top schools in delhi ncr
Top schools in delhi ncrTop schools in delhi ncr
Top schools in delhi ncrEdhole.com
 
Wavelets and Other Adaptive Methods
Wavelets and Other Adaptive MethodsWavelets and Other Adaptive Methods
Wavelets and Other Adaptive MethodsKamrul Hasan
 
4 stochastic processes
4 stochastic processes4 stochastic processes
4 stochastic processesSolo Hermelin
 

Similaire à Voronoy Story (20)

Quick run through on classical mechancis and quantum mechanics
Quick run through on classical mechancis and quantum mechanics Quick run through on classical mechancis and quantum mechanics
Quick run through on classical mechancis and quantum mechanics
 
Calculus of variations
Calculus of variationsCalculus of variations
Calculus of variations
 
Effective Field Theories in the Quest for BSM Physics
Effective Field Theories in the Quest for BSM PhysicsEffective Field Theories in the Quest for BSM Physics
Effective Field Theories in the Quest for BSM Physics
 
lec02.pdf
lec02.pdflec02.pdf
lec02.pdf
 
Oscillations
OscillationsOscillations
Oscillations
 
Cs jog
Cs jogCs jog
Cs jog
 
Ch01 1
Ch01 1Ch01 1
Ch01 1
 
Slides econometrics-2018-graduate-4
Slides econometrics-2018-graduate-4Slides econometrics-2018-graduate-4
Slides econometrics-2018-graduate-4
 
Quantum mechanics
Quantum mechanicsQuantum mechanics
Quantum mechanics
 
Magnetic Pendulum
Magnetic PendulumMagnetic Pendulum
Magnetic Pendulum
 
Admissions in india 2015
Admissions in india 2015Admissions in india 2015
Admissions in india 2015
 
Admission in india 2015
Admission in india 2015Admission in india 2015
Admission in india 2015
 
dalrymple_slides.ppt
dalrymple_slides.pptdalrymple_slides.ppt
dalrymple_slides.ppt
 
Causal Dynamical Triangulations
Causal Dynamical TriangulationsCausal Dynamical Triangulations
Causal Dynamical Triangulations
 
Limits and Continuity of Functions
Limits and Continuity of Functions Limits and Continuity of Functions
Limits and Continuity of Functions
 
Top schools in delhi ncr
Top schools in delhi ncrTop schools in delhi ncr
Top schools in delhi ncr
 
Top schools in delhi ncr
Top schools in delhi ncrTop schools in delhi ncr
Top schools in delhi ncr
 
Wavelets and Other Adaptive Methods
Wavelets and Other Adaptive MethodsWavelets and Other Adaptive Methods
Wavelets and Other Adaptive Methods
 
Wave function
Wave functionWave function
Wave function
 
4 stochastic processes
4 stochastic processes4 stochastic processes
4 stochastic processes
 

Plus de Bruno Levy

On Mesh Intersection: exact computation and efficiency
On Mesh Intersection: exact computation and efficiencyOn Mesh Intersection: exact computation and efficiency
On Mesh Intersection: exact computation and efficiencyBruno Levy
 
Brenier-Monge-Ampère gravity
Brenier-Monge-Ampère gravityBrenier-Monge-Ampère gravity
Brenier-Monge-Ampère gravityBruno Levy
 
SGP 2023 graduate school - A quick journey into geometry processing
SGP 2023 graduate school - A quick journey into geometry processingSGP 2023 graduate school - A quick journey into geometry processing
SGP 2023 graduate school - A quick journey into geometry processingBruno Levy
 
03_spectral_computing.pdf
03_spectral_computing.pdf03_spectral_computing.pdf
03_spectral_computing.pdfBruno Levy
 
04_spectral_applications.pdf
04_spectral_applications.pdf04_spectral_applications.pdf
04_spectral_applications.pdfBruno Levy
 
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)Bruno Levy
 
The joy of computer graphics programming
The joy of computer graphics programmingThe joy of computer graphics programming
The joy of computer graphics programmingBruno Levy
 
Simuler la physique avec un ordinateur
Simuler la physique avec un ordinateurSimuler la physique avec un ordinateur
Simuler la physique avec un ordinateurBruno Levy
 

Plus de Bruno Levy (9)

On Mesh Intersection: exact computation and efficiency
On Mesh Intersection: exact computation and efficiencyOn Mesh Intersection: exact computation and efficiency
On Mesh Intersection: exact computation and efficiency
 
Brenier-Monge-Ampère gravity
Brenier-Monge-Ampère gravityBrenier-Monge-Ampère gravity
Brenier-Monge-Ampère gravity
 
SGP 2023 graduate school - A quick journey into geometry processing
SGP 2023 graduate school - A quick journey into geometry processingSGP 2023 graduate school - A quick journey into geometry processing
SGP 2023 graduate school - A quick journey into geometry processing
 
03_spectral_computing.pdf
03_spectral_computing.pdf03_spectral_computing.pdf
03_spectral_computing.pdf
 
04_spectral_applications.pdf
04_spectral_applications.pdf04_spectral_applications.pdf
04_spectral_applications.pdf
 
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)
Centroidal Voronoi Tessellations for Graphs (Eurographics 2012)
 
Igrv2017
Igrv2017Igrv2017
Igrv2017
 
The joy of computer graphics programming
The joy of computer graphics programmingThe joy of computer graphics programming
The joy of computer graphics programming
 
Simuler la physique avec un ordinateur
Simuler la physique avec un ordinateurSimuler la physique avec un ordinateur
Simuler la physique avec un ordinateur
 

Dernier

Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verifiedConnaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verifiedDelhi Call girls
 
Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Silpa
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfSumit Kumar yadav
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and ClassificationsAreesha Ahmad
 
IDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicineIDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicinesherlingomez2
 
Introduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptxIntroduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptxBhagirath Gogikar
 
Unit5-Cloud.pptx for lpu course cse121 o
Unit5-Cloud.pptx for lpu course cse121 oUnit5-Cloud.pptx for lpu course cse121 o
Unit5-Cloud.pptx for lpu course cse121 oManavSingh202607
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Joonhun Lee
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryAlex Henderson
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)Areesha Ahmad
 
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Silpa
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptxAlMamun560346
 
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.Nitya salvi
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPirithiRaju
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPirithiRaju
 
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptxSCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptxRizalinePalanog2
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPirithiRaju
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flyPRADYUMMAURYA1
 

Dernier (20)

Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verifiedConnaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
Connaught Place, Delhi Call girls :8448380779 Model Escorts | 100% verified
 
Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdf
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and Classifications
 
Site Acceptance Test .
Site Acceptance Test                    .Site Acceptance Test                    .
Site Acceptance Test .
 
IDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicineIDENTIFICATION OF THE LIVING- forensic medicine
IDENTIFICATION OF THE LIVING- forensic medicine
 
Introduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptxIntroduction,importance and scope of horticulture.pptx
Introduction,importance and scope of horticulture.pptx
 
Unit5-Cloud.pptx for lpu course cse121 o
Unit5-Cloud.pptx for lpu course cse121 oUnit5-Cloud.pptx for lpu course cse121 o
Unit5-Cloud.pptx for lpu course cse121 o
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)
 
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
 
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdf
 
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptxSCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
SCIENCE-4-QUARTER4-WEEK-4-PPT-1 (1).pptx
 
Clean In Place(CIP).pptx .
Clean In Place(CIP).pptx                 .Clean In Place(CIP).pptx                 .
Clean In Place(CIP).pptx .
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
 

Voronoy Story

  • 1. Remeshing Variational methods Bruno Lévy ALICE Géométrie & Lumière CENTRE INRIA Nancy Grand-Est
  • 2. OVERVIEW Part. 1. Introduction - Motivations Part. 2. Blowing Bubbles Part. 3. Least Action (the power of power diagrams) Conclusions
  • 10. Voronoi diagram: Vor(xi) = { x | d2(x,xi) < d2(x,xj) } Chap. 2. Centroidal Voronoi Tesselation
  • 12. Chap. 2. Centroidal Voronoi Tesselation The classical method: Lloyd algorithm = gradient descent
  • 13. Chap. 2. Centroidal Voronoi Tesselation The classical method: Lloyd algorithm = gradient descent Which objective function ?
  • 14. Chap. 2. Centroidal Voronoi Tesselation The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) F= ∫Vor(i) 2 dxxi - x i
  • 15. Chap. 2. Centroidal Voronoi Tesselation F= ∫Vor(i) 2 dxxi - x i The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ?
  • 16. Chap. 2. Centroidal Voronoi Tesselation F= ∫Vor(i) 2 dxxi - x i = ∫Inf i 2 dxxi - x The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ?
  • 17. Chap. 2. Centroidal Voronoi Tesselation F= ∫Vor(i) 2 dxxi - x i = ∫Inf i 2 dxxi - x The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ?
  • 18. Chap. 2. Centroidal Voronoi Tesselation F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al] F= ∫Vor(i) 2 dxxi - x i = ∫Inf i 2 dxxi - x The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ?
  • 19. Chap. 2. Centroidal Voronoi Tesselation F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al] F= ∫Vor(i) 2 dxxi - x i = ∫Inf i 2 dxxi - x The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ? Volume of Vor(i)
  • 20. Chap. 2. Centroidal Voronoi Tesselation F|xi = 2 mi (xi - gi) [Iri et.al], [Du et.al] F= ∫Vor(i) 2 dxxi - x i = ∫Inf i 2 dxxi - x The classical method: Lloyd algorithm = gradient descent Which objective function ? This one (quantization noise power) Why Lloyd works ? What is the gradient of F ? Volume of Vor(i) Centroid of Vor(i)
  • 21. Chap. 2. Centroidal Voronoi Tesselation The classical method: Lloyd algorithm = gradient descent F= ∫Vor(i) 2 dxxi - x i F is smooth (C2) [Liu, Wang, L, Sun, Yan, Lu, Yang 2009]
  • 22. Chap. 2. Centroidal Voronoi Tesselation The classical method: Lloyd algorithm = gradient descent F= ∫Vor(i) 2 dxxi - x i F is smooth (C2) [Liu, Wang, L, Sun, Yan, Lu, Yang 2009]
  • 23. [Yan, L, Liu, Sun and Wang 2009]
  • 24. Chap. 2. Centroidal Voronoi Tesselation F= ∫Vor(i) 2 dxxi - x i Blowing square bubbles
  • 25. Chap. 2. Centroidal Voronoi Tesselation p=2 p=4 p=8 …. F= ∫Vor(i) 2 dxxi - x i p
  • 27. Chap. 2. Centroidal Voronoi Tesselation [L, Liu, 2010]
  • 28. Chap. 2. Centroidal Voronoi Tesselation [L, Liu, 2010] [Ray, Sokolov, L, 2015, 2016] (completely different method: direction fields + global parameterization)
  • 29. Least Action Power Diagrams and Optimal Transport 3
  • 30. Yann Brenier The polar factorization theorem (Brenier Transport) Each time the Laplace operator is used in a PDE, it can be replaced with the Monte-Ampère operator, and then interesting things occur
  • 31. Euler Lagrange The Least Action Principle Hamilton, Legendre, Maupertuis Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system Short summary of the 1st chapter of Landau,Lifshitz Course of Theoretical Physics
  • 32. Euler Lagrange The Least Action Principle Hamilton, Legendre, Maupertuis Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system position
  • 33. Euler Lagrange The Least Action Principle Hamilton, Legendre, Maupertuis Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system position speed
  • 34. Euler Lagrange The Least Action Principle Hamilton, Legendre, Maupertuis Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system position speed time
  • 35. Euler Lagrange ∫t1 t2 L(x,x,t) dt The Least Action Principle Hamilton, Legendre, Maupertuis Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system Axiom 2: The movement (time evolution) of the physical system minimizes the following integral
  • 36. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system Axiom 2: The movement (time evolution) of the physical system minimizes the following integral
  • 37. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists a function L(x,x,t) that describes the state of a physical system Axiom 2: The movement (time evolution) of the physical system minimizes the following integral Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = t=0 t=1
  • 38. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t
  • 39. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte
  • 40. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy
  • 41. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum
  • 42. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum Isotropy of space => Preservation of angular momentum
  • 43. ∫t1 t2 L(x,x,t) dt The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum Isotropy of space => Preservation of angular momentum Preserved quantities Integrals of Motion Noeter theorem
  • 44. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum Isotropy of space => Preservation of angular momentum Free particle: Theorem 3: v = cte (Newton law I)
  • 45. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum Isotropy of space => Preservation of angular momentum Free particle: Theorem 3: v = cte (Newton law I) Expression of the Lagrangian: L = ½ m v2
  • 46. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Theorem 2: ∂L ∂x x - L = cte Homogeneity of time => Preservation of energy Homogeneity of space => Preservation of momentum Isotropy of space => Preservation of angular momentum Free particle: Theorem 3: v = cte (Newton law I) Expression of the Lagrangian: L = ½ m v2 Expression of the Energy: E = ½ m v2
  • 47. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Particle in a field: Expression of the Lagrangian: L = ½ m v2 – U(x) Free particle: Theorem 3: v = cte (Newton law I) Expression of the Lagrangian: L = ½ m v2 Expression of the Energy: E = ½ m v2
  • 48. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Particle in a field: Expression of the Lagrangian: L = ½ m v2 – U(x) Expression of the Energy: E = ½ m v2 + U(x) Free particle: Theorem 3: v = cte (Newton law I) Expression of the Lagrangian: L = ½ m v2 Expression of the Energy: E = ½ m v2
  • 49. The Least Action Principle Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. change of Gallileo frame + hom. + isotrop. : x’ t’ = x+vt t Particle in a field: Expression of the Lagrangian: L = ½ m v2 – U(x) Expression of the Energy: E = ½ m v2 + U(x) Theorem 4: mx = - U (Newton law II) Free particle: Theorem 3: v = cte (Newton law I) Expression of the Lagrangian: L = ½ m v2 Expression of the Energy: E = ½ m v2
  • 50. Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. Lorentz change of frame x’ t’ = (x-vt) x γ (t – vx/c2) x γ γ = 1 / ( 1 – v2 / c2) The Least Action Principle (relativistic setting, just for fun) √
  • 51. The Least Action Principle (relativistic setting, just for fun) Axiom 1: There exists L Axiom 2: The movement minimizes ∫ L Theorem 1: (Lagrange equation): ∂L ∂x d dt ∂L ∂x = Axiom 3: Invariance w.r.t. Lorentz change of frame x’ t’ = (x-vt) x γ (t – vx/c2) x γ Theorem 5: E = ½ γ m v2 + mc2 γ = 1 / ( 1 – v2 / c2)√
  • 52. The Least Action Principle (quantum physics setting, just for fun)
  • 55. FluidsLagrange point of view ρ nb particles per square Euler point of view
  • 56. FluidsLagrange point of view ρ(x,y,t) nb particles per square Euler point of view v(x,y,t) speed of the particle under grid point (x,y) at time t
  • 57. Fluids ρ(x,y,t) nb particles per square Euler point of view v(x,y,t) speed of the particle under grid point (x,y) at time t Q1: how to compute the acceleration of the particles from v(x,y,t) ? dv dt = ∂v v. v ∂t + Q2: incompressible fluids ? div(v) = 0 Q3: mass preservation ? d ρ dt = - div(ρv) (Continuity equation)
  • 58. Fluids Start with Lagrange coordinates: particle trajectories: X(t,x) Minimize Action: ∫t1 t2 1/2 ∫Ω ∂X (t,x) ∂t dxdt (ρ = cte) s.t. X satisfies mass preservation (X is measure-preserving, more on this later) 2
  • 59. Fluids Start with Lagrange coordinates: particle trajectories: X(t,x) Minimize Action: ∫t1 t2 1/2 ∫Ω ∂X (t,x) ∂t dxdt (ρ = cte) s.t. X satisfies mass preservation (X is measure-preserving, more on this later) Acceleration of the particle under the grid 2
  • 60. Fluids Start with Lagrange coordinates: particle trajectories: X(t,x) Minimize Action: ∫t1 t2 1/2 ∫Ω ∂X (t,x) ∂t dxdt (ρ = cte) s.t. X satisfies mass preservation (X is measure-preserving, more on this later) The Lagrange multiplier for the constraint = pressure 2
  • 61. ? T Fluids – Benamou Brenier ρ1 ρ2
  • 62. ? T Fluids – Benamou Brenier Minimize A(ρ,v) = ∫t1 t2 (t2-t1) ∫Ω ρ(x,t) ||v(t,x)||2 dxdt s.t. ρ(t1,.) = ρ1 ; ρ(t2,.) = ρ2 ; d ρ dt = - div(ρv) ρ1 ρ2
  • 63. ? T Fluids – Benamou Brenier Minimize A(ρ,v) = ∫t1 t2 (t2-t1) ∫Ω ρ(x,t) ||v(t,x)||2 dxdt s.t. ρ(t1,.) = ρ1 ; ρ(t2,.) = ρ2 ; d ρ dt = - div(ρv) ρ1 ρ2 Minimize C(T) = ∫Ω || x – T(x) ||2 dx s.t. T is measure-preserving ρ1(x)
  • 64. Part. 3 Optimal Transport – Monge problem A map T is a transport map between μ and ν if μ(T-1(B)) = ν(B) for any Borel subset B of Y (X;μ) (Y;ν)
  • 65. Part. 3 Optimal Transport – Monge problem A map T is a transport map between μ and ν if μ(T-1(B)) = ν(B) for any Borel subset B B (X;μ) (Y;ν)
  • 66. Part. 3 Optimal Transport – Monge problem A map T is a transport map between μ and ν if μ(T-1(B)) = ν(B) for any Borel subset B B T-1(B) (X;μ) (Y;ν)
  • 67. Part. 3 Optimal Transport – Monge problem A map T is a transport map between μ and ν if μ(T-1(B)) = ν(B) for any Borel subset B (or ν = T#μ the pushforward of μ) (X;μ) (Y;ν)
  • 68. Part. 3 Optimal Transport – Monge problem Monge problem (1787): Find a transport map T that minimizes C(T) = ∫X || x – T(x) ||2 dμ(x) (X;μ) (Y;ν)
  • 69. Part. 3 Optimal Transport – Kantorovich Monge problem: Find a transport map T that minimizes C(T) = ∫X || x – T(x) ||2 dμ(x) Kantorovich problem (1942): Find a measure γ defined on X x Y such that ∫x in X dγ(x,y) = dν(y) and ∫y in Y dγ(x,y) = dμ(x) that minimizes ∫∫X x Y || x – y ||2 dγ(x,y)
  • 70. Part. 3 Optimal Transport – Kantorovich Transport plan – example 1/2 : translation of a segment
  • 71. Part. 3 Optimal Transport – Kantorovich Transport plan – example 1/2 : translation of a segment
  • 72. Part. 3 Optimal Transport – Kantorovich
  • 73. Part. 3 Optimal Transport – Dual of Kantorovich Dual formulation of Kantorovich problem: Find a c-concave function ψ that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform)
  • 74. Part. 3 Optimal Transport – Dual of Kantorovich Dual formulation of Kantorovich problem: Find a c-concave function ψ that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform) What about our initial problem ? T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) ) {{grad ψ(x) with ψ(x) := (½ x2- ψ(x)) [Brenier]
  • 75. Part. 3 Optimal Transport – Dual of Kantorovich Dual formulation of Kantorovich problem: Find a c-concave function ψ that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform) What about our initial problem ? T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) ) {{grad ψ(x) with ψ(x) := (½ x2- ψ(x)) When μ and ν have a density u and v, (H ψ(x)). v(grad ψ(x)) = u(x) Monge-Ampère equation for all borel set A, ∫A dμ = ∫T(A) (|JT|) dν = ∫T(A) (H ψ ) dν [Brenier]
  • 76. Part. 3 Optimal Transport – Dual of Kantorovich Dual formulation of Kantorovich problem: Find a c-concave function ψ that maximizes ∫X ψ(x)dμ + ∫Y ψc(y)dν ψc(y) = inf x [ c(x,y) - ψ(x) ]where: (Legendre transform) What about our initial problem ? T(x) = x – grad ψ(x) = grad (½ x2- ψ(x) ) {{grad ψ(x) with ψ(x) := (½ x2- ψ(x)) When μ and ν have a density u and v, (H ψ(x)). v(grad ψ(x)) = u(x) Monge-Ampère equation for all borel set A, ∫A dμ = ∫T(A) (|JT|) dν = ∫T(A) (H ψ ) dν [Brenier]
  • 77. Part. 3 Optimal Transport – semi-discrete ∫X ψc (x)dμ + ∫Y ψ(y)dν Sup ψ Є ψc (DMK) (X;μ) (Y;ν)
  • 78. Part. 3 Optimal Transport – semi-discrete (X;μ) (Y;ν) ∑j ψ(yj) vj ∫X ψc (x)dμ + ∫Y ψ(y)dν Sup ψ Є ψc (DMK)
  • 79. Part. 3 Optimal Transport – semi-discrete ∫X ψc (x)dμ + ∫Y ψ(y)dν Sup ψ Є ψc (DMK) ∑j ψ(yj) vj
  • 80. Part. 3 Optimal Transport – semi-discrete ∫X ψc (x)dμ + ∫Y ψ(y)dν Sup ψ Є ψc (DMK) ∫X inf yj ЄY [ || x – yj ||2 - ψ(yj) ] dμ ∑j ψ(yj) vj
  • 81. Part. 3 Optimal Transport – semi-discrete ∫X ψc (x)dμ + ∫Y ψ(y)dν Sup ψ Є ψc (DMK) ∑j ∫Lagψ(yj) || x – yj ||2 - ψ(yj) dμ ∫X inf yj ЄY [ || x – yj ||2 - ψ(yj) ] dμ ∑j ψ(yj) vj
  • 82. Power diagram: Pow(xi) = { x | d2(x,xi) – ψi < d2(x,xj) – ψj } Voronoi diagram: Vor(xi) = { x | d2(x,xi) < d2(x,xj) } Part. 3 Power Diagrams
  • 83. Part. 3 Power Diagrams
  • 84. Part. 3 Optimal Transport Theorem: (direct consequence of MK duality alternative proof in [Aurenhammer, Hoffmann, Aronov 98] ): Given a measure μ with density, a set of points (yj), a set of positive coefficients vj such that ∑ vj =∫ dμ(x), it is possible to find the weights W = [ψ(y1) ψ(y2) … ψ(ym)] such that the map TS W is the unique optimal transport map between μ and ν = ∑ vj δ(yj)
  • 85. Part. 3 Optimal Transport Optimal transport Centroidal Voronoi Tesselation [L - A numerical Algorithm for L2 semi-discrete OT (ESAIM M2AN 2015)]
  • 86. Part. 3 Power Diagrams & Transport
  • 87. Part. 3 Power Diagrams & Transport
  • 88. Part. 3 Power Diagrams & Transport
  • 89. Part. 3 Power Diagrams & Transport
  • 90. Part. 3 Power Diagrams & Transport
  • 91. Part. 3 Power Diagrams & Transport
  • 92. Part. 3 Power Diagrams & Transport hi
  • 93. Part. 3 Power Diagrams & Transport
  • 94. Part. 3 Power Diagrams & Transport
  • 95. Part. 3 Power Diagrams & Transport
  • 96. Part. 3 Power Diagrams & Transport
  • 97. Part. 3 Power Diagrams and Transport
  • 98. Part. 3 Optimal Transport – 2D examples Numerical Experiment: A disk becomes two disks
  • 99. Part. 3 Optimal Transport – 3D examples Numerical Experiment: A sphere becomes two spheres
  • 100. Part. 3 Optimal Transport – 3D examples Numerical Experiment: Armadillo to sphere
  • 101. Part. 3 Optimal Transport – 3D examples Numerical Experiment: Other examples
  • 102. Part. 3 Optimal Transport – 3D examples Numerical Experiment: Varying density
  • 103. Part. 1 Optimal Transport [Schwartzburg et.al, SIGGRAPH 2014]
  • 104. Part. 1 Optimal Transport Optimal transport E.U.R. let there be light !
  • 105. Part. 1 Optimal Transport Optimal transport E.U.R. let there be light ! The millenium simulation project, Max Planck Institute fur Astrophysik pc/h : parsec (= 3.2 light years)
  • 106. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n))
  • 107. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n)) 2015: 3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
  • 108. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n)) 2015: 3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015]
  • 109. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n)) 2015: 3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015] 2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D: 1 million Dirac masses in 240 seconds
  • 110. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n)) 2015: 3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015] 2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D: 1 million Dirac masses in 240 seconds 10 million Dirac masses in 90 minutes
  • 111. Part. 4 Optimal Transport – 3D examples Numerical Experiment: Performances 2002: several hours of supercomputer time were needed for computing OT with a few thousand Dirac masses, with a combinatorial algorithm in O(n2log(n)) 2015: 3D version of [Mérigot] (multilevel + BFGS) + several tricks [L 2015] 2016: Damped Newton [Mérigot, Thibert] + several tricks [L] for 3D: 1 million Dirac masses in 240 seconds 10 million Dirac masses in 90 minutes Semi-discrete OT is now scalable ! (new tool in Num. Ana. Toolbox)
  • 112. Other topics •Euler equation in more complicated setting: [Merigot & Mirebeau, Merigot & Galouet] •Using semi-discrete OT to solve other PDEs [Benamou, Carlier, Merigot , Oudet] •Faster solvers
  • 113. Online resources Source code: alice.loria.fr/software/geogram Demos: www.loria.fr/~levy/GEOGRAM/ www.loria.fr/~levy/GLUP/ L., A numerical algorithm for semi-discrete L2 OT in 3D, ESAIM Math. Modeling and Analysis, 2015 Video (presentation with the proofs) www.loria.fr/~levy (Ecole d été calcul des variations 2015) New projects: MAGA Q. Mérigot - ANR EXPLORAGRAM (Inria) with E. Schwindt, Q. Mérigot and J.-D. Benamou