Point Cloud Library (PCL) 1.15.0
Loading...
Searching...
No Matches
transformation_estimation_dual_quaternion.hpp
1/*
2 * Software License Agreement (BSD License)
3 *
4 * Point Cloud Library (PCL) - www.pointclouds.org
5 * Copyright (c) 2010, Willow Garage, Inc.
6 * Copyright (c) 2012-, Open Perception, Inc.
7 *
8 * All rights reserved.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 *
14 * * Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * * Redistributions in binary form must reproduce the above
17 * copyright notice, this list of conditions and the following
18 * disclaimer in the documentation and/or other materials provided
19 * with the distribution.
20 * * Neither the name of the copyright holder(s) nor the names of its
21 * contributors may be used to endorse or promote products derived
22 * from this software without specific prior written permission.
23 *
24 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
25 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
26 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
27 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
28 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
29 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
30 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
31 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
32 * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
33 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
34 * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
35 * POSSIBILITY OF SUCH DAMAGE.
36 *
37 *
38 */
39
40#ifndef PCL_REGISTRATION_TRANSFORMATION_ESTIMATION_DQ_HPP_
41#define PCL_REGISTRATION_TRANSFORMATION_ESTIMATION_DQ_HPP_
42
43#include <pcl/common/eigen.h>
44
45#include <Eigen/Eigenvalues> // for EigenSolver
46
47namespace pcl {
48
49namespace registration {
50
51template <typename PointSource, typename PointTarget, typename Scalar>
52inline void
55 const pcl::PointCloud<PointTarget>& cloud_tgt,
56 Matrix4& transformation_matrix) const
57{
58 const auto nr_points = cloud_src.size();
59 if (cloud_tgt.size() != nr_points) {
60 PCL_ERROR(
61 "[pcl::TransformationEstimationDualQuaternion::estimateRigidTransformation] "
62 "Number or points in source (%zu) differs than target (%zu)!\n",
63 static_cast<std::size_t>(nr_points),
64 static_cast<std::size_t>(cloud_tgt.size()));
65 return;
66 }
67
68 ConstCloudIterator<PointSource> source_it(cloud_src);
69 ConstCloudIterator<PointTarget> target_it(cloud_tgt);
70 estimateRigidTransformation(source_it, target_it, transformation_matrix);
71}
72
73template <typename PointSource, typename PointTarget, typename Scalar>
74void
77 const pcl::Indices& indices_src,
78 const pcl::PointCloud<PointTarget>& cloud_tgt,
79 Matrix4& transformation_matrix) const
80{
81 if (indices_src.size() != cloud_tgt.size()) {
82 PCL_ERROR("[pcl::TransformationDQ::estimateRigidTransformation] Number or points "
83 "in source (%zu) differs than target (%zu)!\n",
84 indices_src.size(),
85 static_cast<std::size_t>(cloud_tgt.size()));
86 return;
87 }
88
89 ConstCloudIterator<PointSource> source_it(cloud_src, indices_src);
90 ConstCloudIterator<PointTarget> target_it(cloud_tgt);
91 estimateRigidTransformation(source_it, target_it, transformation_matrix);
92}
93
94template <typename PointSource, typename PointTarget, typename Scalar>
95inline void
98 const pcl::Indices& indices_src,
99 const pcl::PointCloud<PointTarget>& cloud_tgt,
100 const pcl::Indices& indices_tgt,
101 Matrix4& transformation_matrix) const
102{
103 if (indices_src.size() != indices_tgt.size()) {
104 PCL_ERROR(
105 "[pcl::TransformationEstimationDualQuaternion::estimateRigidTransformation] "
106 "Number or points in source (%lu) differs than target (%lu)!\n",
107 indices_src.size(),
108 indices_tgt.size());
109 return;
110 }
111
112 ConstCloudIterator<PointSource> source_it(cloud_src, indices_src);
113 ConstCloudIterator<PointTarget> target_it(cloud_tgt, indices_tgt);
114 estimateRigidTransformation(source_it, target_it, transformation_matrix);
115}
116
117template <typename PointSource, typename PointTarget, typename Scalar>
118void
121 const pcl::PointCloud<PointTarget>& cloud_tgt,
122 const pcl::Correspondences& correspondences,
123 Matrix4& transformation_matrix) const
124{
125 ConstCloudIterator<PointSource> source_it(cloud_src, correspondences, true);
126 ConstCloudIterator<PointTarget> target_it(cloud_tgt, correspondences, false);
127 estimateRigidTransformation(source_it, target_it, transformation_matrix);
128}
129
130template <typename PointSource, typename PointTarget, typename Scalar>
131inline void
135 Matrix4& transformation_matrix) const
136{
137 const int npts = static_cast<int>(source_it.size());
138
139 transformation_matrix.setIdentity();
140
141 // dual quaternion optimization
142 Eigen::Matrix<double, 4, 4> C1 = Eigen::Matrix<double, 4, 4>::Zero();
143 Eigen::Matrix<double, 4, 4> C2 = Eigen::Matrix<double, 4, 4>::Zero();
144 double* c1 = C1.data();
145 double* c2 = C2.data();
146
147 for (int i = 0; i < npts; ++i) {
148 const PointSource& a = *source_it;
149 const PointTarget& b = *target_it;
150 const double axbx = a.x * b.x;
151 const double ayby = a.y * b.y;
152 const double azbz = a.z * b.z;
153 const double axby = a.x * b.y;
154 const double aybx = a.y * b.x;
155 const double axbz = a.x * b.z;
156 const double azbx = a.z * b.x;
157 const double aybz = a.y * b.z;
158 const double azby = a.z * b.y;
159 c1[0] += axbx - azbz - ayby;
160 c1[5] += ayby - azbz - axbx;
161 c1[10] += azbz - axbx - ayby;
162 c1[15] += axbx + ayby + azbz;
163 c1[1] += axby + aybx;
164 c1[2] += axbz + azbx;
165 c1[3] += aybz - azby;
166 c1[6] += azby + aybz;
167 c1[7] += azbx - axbz;
168 c1[11] += axby - aybx;
169
170 c2[1] += a.z + b.z;
171 c2[2] -= a.y + b.y;
172 c2[3] += a.x - b.x;
173 c2[6] += a.x + b.x;
174 c2[7] += a.y - b.y;
175 c2[11] += a.z - b.z;
176 ++source_it;
177 ++target_it;
178 }
179
180 c1[4] = c1[1];
181 c1[8] = c1[2];
182 c1[9] = c1[6];
183 c1[12] = c1[3];
184 c1[13] = c1[7];
185 c1[14] = c1[11];
186 c2[4] = -c2[1];
187 c2[8] = -c2[2];
188 c2[12] = -c2[3];
189 c2[9] = -c2[6];
190 c2[13] = -c2[7];
191 c2[14] = -c2[11];
192
193 C1 *= -2.0;
194 C2 *= 2.0;
195
196 const Eigen::Matrix<double, 4, 4> A =
197 (0.25 / static_cast<double>(npts)) * C2.transpose() * C2 - C1;
198
199 const Eigen::EigenSolver<Eigen::Matrix<double, 4, 4>> es(A);
200
201 ptrdiff_t i;
202 es.eigenvalues().real().maxCoeff(&i);
203 const Eigen::Matrix<double, 4, 1> qmat = es.eigenvectors().col(i).real();
204 const Eigen::Matrix<double, 4, 1> smat =
205 -(0.5 / static_cast<double>(npts)) * C2 * qmat;
206
207 const Eigen::Quaternion<double> q(qmat(3), qmat(0), qmat(1), qmat(2));
208 const Eigen::Quaternion<double> s(smat(3), smat(0), smat(1), smat(2));
209
210 const Eigen::Quaternion<double> t = s * q.conjugate();
211
212 const Eigen::Matrix<double, 3, 3> R(q.toRotationMatrix());
213
214 for (int i = 0; i < 3; ++i)
215 for (int j = 0; j < 3; ++j)
216 transformation_matrix(i, j) = R(i, j);
217
218 transformation_matrix(0, 3) = -t.x();
219 transformation_matrix(1, 3) = -t.y();
220 transformation_matrix(2, 3) = -t.z();
221}
222
223} // namespace registration
224} // namespace pcl
225
226#endif /* PCL_REGISTRATION_TRANSFORMATION_ESTIMATION_DQ_HPP_ */
Iterator class for point clouds with or without given indices.
std::size_t size() const
Size of the range the iterator is going through.
PointCloud represents the base class in PCL for storing collections of 3D points.
std::size_t size() const
void estimateRigidTransformation(const pcl::PointCloud< PointSource > &cloud_src, const pcl::PointCloud< PointTarget > &cloud_tgt, Matrix4 &transformation_matrix) const override
Estimate a rigid rotation transformation between a source and a target point cloud using dual quatern...
typename TransformationEstimation< PointSource, PointTarget, Scalar >::Matrix4 Matrix4
std::vector< pcl::Correspondence, Eigen::aligned_allocator< pcl::Correspondence > > Correspondences
IndicesAllocator<> Indices
Type used for indices in PCL.
Definition types.h:133