dust3d/third_party/libigl/include/igl/arap.cpp

313 lines
8.5 KiB
C++

// This file is part of libigl, a simple c++ geometry processing library.
//
// Copyright (C) 2013 Alec Jacobson <alecjacobson@gmail.com>
//
// This Source Code Form is subject to the terms of the Mozilla Public License
// v. 2.0. If a copy of the MPL was not distributed with this file, You can
// obtain one at http://mozilla.org/MPL/2.0/.
#include "arap.h"
#include "colon.h"
#include "cotmatrix.h"
#include "massmatrix.h"
#include "group_sum_matrix.h"
#include "covariance_scatter_matrix.h"
#include "speye.h"
#include "mode.h"
#include "project_isometrically_to_plane.h"
#include "slice.h"
#include "arap_rhs.h"
#include "repdiag.h"
#include "columnize.h"
#include "fit_rotations.h"
#include <cassert>
#include <iostream>
template <
typename DerivedV,
typename DerivedF,
typename Derivedb>
IGL_INLINE bool igl::arap_precomputation(
const Eigen::PlainObjectBase<DerivedV> & V,
const Eigen::PlainObjectBase<DerivedF> & F,
const int dim,
const Eigen::PlainObjectBase<Derivedb> & b,
ARAPData & data)
{
using namespace std;
using namespace Eigen;
typedef typename DerivedV::Scalar Scalar;
// number of vertices
const int n = V.rows();
data.n = n;
assert((b.size() == 0 || b.maxCoeff() < n) && "b out of bounds");
assert((b.size() == 0 || b.minCoeff() >=0) && "b out of bounds");
// remember b
data.b = b;
//assert(F.cols() == 3 && "For now only triangles");
// dimension
//const int dim = V.cols();
assert((dim == 3 || dim ==2) && "dim should be 2 or 3");
data.dim = dim;
//assert(dim == 3 && "Only 3d supported");
// Defaults
data.f_ext = MatrixXd::Zero(n,data.dim);
assert(data.dim <= V.cols() && "solve dim should be <= embedding");
bool flat = (V.cols() - data.dim)==1;
DerivedV plane_V;
DerivedF plane_F;
typedef SparseMatrix<Scalar> SparseMatrixS;
SparseMatrixS ref_map,ref_map_dim;
if(flat)
{
project_isometrically_to_plane(V,F,plane_V,plane_F,ref_map);
repdiag(ref_map,dim,ref_map_dim);
}
const PlainObjectBase<DerivedV>& ref_V = (flat?plane_V:V);
const PlainObjectBase<DerivedF>& ref_F = (flat?plane_F:F);
SparseMatrixS L;
cotmatrix(V,F,L);
ARAPEnergyType eff_energy = data.energy;
if(eff_energy == ARAP_ENERGY_TYPE_DEFAULT)
{
switch(F.cols())
{
case 3:
if(data.dim == 3)
{
eff_energy = ARAP_ENERGY_TYPE_SPOKES_AND_RIMS;
}else
{
eff_energy = ARAP_ENERGY_TYPE_ELEMENTS;
}
break;
case 4:
eff_energy = ARAP_ENERGY_TYPE_ELEMENTS;
break;
default:
assert(false);
}
}
// Get covariance scatter matrix, when applied collects the covariance
// matrices used to fit rotations to during optimization
covariance_scatter_matrix(ref_V,ref_F,eff_energy,data.CSM);
if(flat)
{
data.CSM = (data.CSM * ref_map_dim.transpose()).eval();
}
assert(data.CSM.cols() == V.rows()*data.dim);
// Get group sum scatter matrix, when applied sums all entries of the same
// group according to G
SparseMatrix<double> G_sum;
if(data.G.size() == 0)
{
if(eff_energy == ARAP_ENERGY_TYPE_ELEMENTS)
{
speye(F.rows(),G_sum);
}else
{
speye(n,G_sum);
}
}else
{
// groups are defined per vertex, convert to per face using mode
if(eff_energy == ARAP_ENERGY_TYPE_ELEMENTS)
{
Eigen::Matrix<int,Eigen::Dynamic,1> GG;
MatrixXi GF(F.rows(),F.cols());
for(int j = 0;j<F.cols();j++)
{
Matrix<int,Eigen::Dynamic,1> GFj;
slice(data.G,F.col(j),GFj);
GF.col(j) = GFj;
}
mode<int>(GF,2,GG);
data.G=GG;
}
//printf("group_sum_matrix()\n");
group_sum_matrix(data.G,G_sum);
}
SparseMatrix<double> G_sum_dim;
repdiag(G_sum,data.dim,G_sum_dim);
assert(G_sum_dim.cols() == data.CSM.rows());
data.CSM = (G_sum_dim * data.CSM).eval();
arap_rhs(ref_V,ref_F,data.dim,eff_energy,data.K);
if(flat)
{
data.K = (ref_map_dim * data.K).eval();
}
assert(data.K.rows() == data.n*data.dim);
SparseMatrix<double> Q = (-L).eval();
if(data.with_dynamics)
{
const double h = data.h;
assert(h != 0);
SparseMatrix<double> M;
massmatrix(V,F,MASSMATRIX_TYPE_DEFAULT,data.M);
const double dw = (1./data.ym)*(h*h);
SparseMatrix<double> DQ = dw * 1./(h*h)*data.M;
Q += DQ;
// Dummy external forces
data.f_ext = MatrixXd::Zero(n,data.dim);
data.vel = MatrixXd::Zero(n,data.dim);
}
return min_quad_with_fixed_precompute(
Q,b,SparseMatrix<double>(),true,data.solver_data);
}
template <
typename Derivedbc,
typename DerivedU>
IGL_INLINE bool igl::arap_solve(
const Eigen::PlainObjectBase<Derivedbc> & bc,
ARAPData & data,
Eigen::PlainObjectBase<DerivedU> & U)
{
using namespace Eigen;
using namespace std;
assert(data.b.size() == bc.rows());
if(bc.size() > 0)
{
assert(bc.cols() == data.dim && "bc.cols() match data.dim");
}
const int n = data.n;
int iter = 0;
if(U.size() == 0)
{
// terrible initial guess.. should at least copy input mesh
#ifndef NDEBUG
cerr<<"arap_solve: Using terrible initial guess for U. Try U = V."<<endl;
#endif
U = MatrixXd::Zero(data.n,data.dim);
}else
{
assert(U.cols() == data.dim && "U.cols() match data.dim");
}
// changes each arap iteration
MatrixXd U_prev = U;
// doesn't change for fixed with_dynamics timestep
MatrixXd U0;
if(data.with_dynamics)
{
U0 = U_prev;
}
while(iter < data.max_iter)
{
U_prev = U;
// enforce boundary conditions exactly
for(int bi = 0;bi<bc.rows();bi++)
{
U.row(data.b(bi)) = bc.row(bi);
}
const auto & Udim = U.replicate(data.dim,1);
assert(U.cols() == data.dim);
// As if U.col(2) was 0
MatrixXd S = data.CSM * Udim;
// THIS NORMALIZATION IS IMPORTANT TO GET SINGLE PRECISION SVD CODE TO WORK
// CORRECTLY.
S /= S.array().abs().maxCoeff();
const int Rdim = data.dim;
MatrixXd R(Rdim,data.CSM.rows());
if(R.rows() == 2)
{
fit_rotations_planar(S,R);
}else
{
fit_rotations(S,true,R);
//#ifdef __SSE__ // fit_rotations_SSE will convert to float if necessary
// fit_rotations_SSE(S,R);
//#else
// fit_rotations(S,true,R);
//#endif
}
//for(int k = 0;k<(data.CSM.rows()/dim);k++)
//{
// R.block(0,dim*k,dim,dim) = MatrixXd::Identity(dim,dim);
//}
// Number of rotations: #vertices or #elements
int num_rots = data.K.cols()/Rdim/Rdim;
// distribute group rotations to vertices in each group
MatrixXd eff_R;
if(data.G.size() == 0)
{
// copy...
eff_R = R;
}else
{
eff_R.resize(Rdim,num_rots*Rdim);
for(int r = 0;r<num_rots;r++)
{
eff_R.block(0,Rdim*r,Rdim,Rdim) =
R.block(0,Rdim*data.G(r),Rdim,Rdim);
}
}
MatrixXd Dl;
if(data.with_dynamics)
{
assert(data.M.rows() == n &&
"No mass matrix. Call arap_precomputation if changing with_dynamics");
const double h = data.h;
assert(h != 0);
//Dl = 1./(h*h*h)*M*(-2.*V0 + Vm1) - fext;
// data.vel = (V0-Vm1)/h
// h*data.vel = (V0-Vm1)
// -h*data.vel = -V0+Vm1)
// -V0-h*data.vel = -2V0+Vm1
const double dw = (1./data.ym)*(h*h);
Dl = dw * (1./(h*h)*data.M*(-U0 - h*data.vel) - data.f_ext);
}
VectorXd Rcol;
columnize(eff_R,num_rots,2,Rcol);
VectorXd Bcol = -data.K * Rcol;
assert(Bcol.size() == data.n*data.dim);
for(int c = 0;c<data.dim;c++)
{
VectorXd Uc,Bc,bcc,Beq;
Bc = Bcol.block(c*n,0,n,1);
if(data.with_dynamics)
{
Bc += Dl.col(c);
}
if(bc.size()>0)
{
bcc = bc.col(c);
}
min_quad_with_fixed_solve(
data.solver_data,
Bc,bcc,Beq,
Uc);
U.col(c) = Uc;
}
iter++;
}
if(data.with_dynamics)
{
// Keep track of velocity for next time
data.vel = (U-U0)/data.h;
}
return true;
}
#ifdef IGL_STATIC_LIBRARY
template bool igl::arap_solve<Eigen::Matrix<double, -1, -1, 0, -1, -1>, Eigen::Matrix<double, -1, -1, 0, -1, -1> >(Eigen::PlainObjectBase<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&, igl::ARAPData&, Eigen::PlainObjectBase<Eigen::Matrix<double, -1, -1, 0, -1, -1> >&);
template bool igl::arap_precomputation<Eigen::Matrix<double, -1, -1, 0, -1, -1>, Eigen::Matrix<int, -1, -1, 0, -1, -1>, Eigen::Matrix<int, -1, 1, 0, -1, 1> >(Eigen::PlainObjectBase<Eigen::Matrix<double, -1, -1, 0, -1, -1> > const&, Eigen::PlainObjectBase<Eigen::Matrix<int, -1, -1, 0, -1, -1> > const&, int, Eigen::PlainObjectBase<Eigen::Matrix<int, -1, 1, 0, -1, 1> > const&, igl::ARAPData&);
#endif