You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
223 lines
5.4 KiB
223 lines
5.4 KiB
/* |
|
* Software License Agreement (BSD License) |
|
* |
|
* Copyright (c) 2009, Willow Garage, Inc. |
|
* All rights reserved. |
|
* |
|
* Redistribution and use in source and binary forms, with or without |
|
* modification, are permitted provided that the following conditions |
|
* are met: |
|
* |
|
* * Redistributions of source code must retain the above copyright |
|
* notice, this list of conditions and the following disclaimer. |
|
* * Redistributions in binary form must reproduce the above |
|
* copyright notice, this list of conditions and the following |
|
* disclaimer in the documentation and/or other materials provided |
|
* with the distribution. |
|
* * Neither the name of Willow Garage, Inc. nor the names of its |
|
* contributors may be used to endorse or promote products derived |
|
* from this software without specific prior written permission. |
|
* |
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
|
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
|
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS |
|
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE |
|
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, |
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, |
|
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
|
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
|
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN |
|
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
|
* POSSIBILITY OF SUCH DAMAGE. |
|
* |
|
*/ |
|
|
|
#include "precomp.hpp" |
|
|
|
// Eigen |
|
#include <Eigen/Core> |
|
|
|
// OpenCV |
|
#include <opencv2/core/eigen.hpp> |
|
#include <opencv2/sfm/numeric.hpp> |
|
#include <opencv2/sfm/projection.hpp> |
|
|
|
// libmv headers |
|
#include "libmv/multiview/projection.h" |
|
|
|
#include <iostream> |
|
|
|
namespace cv |
|
{ |
|
namespace sfm |
|
{ |
|
|
|
template<typename T> |
|
void |
|
homogeneousToEuclidean(const Mat & _X, Mat & _x) |
|
{ |
|
int d = _X.rows - 1; |
|
|
|
const Mat_<T> & X_rows = _X.rowRange(0,d); |
|
const Mat_<T> h = _X.row(d); |
|
|
|
const T * h_ptr = h[0], *h_ptr_end = h_ptr + h.cols; |
|
const T * X_ptr = X_rows[0]; |
|
T * x_ptr = _x.ptr<T>(0); |
|
for (; h_ptr != h_ptr_end; ++h_ptr, ++X_ptr, ++x_ptr) |
|
{ |
|
const T * X_col_ptr = X_ptr; |
|
T * x_col_ptr = x_ptr, *x_col_ptr_end = x_col_ptr + d * _x.step1(); |
|
for (; x_col_ptr != x_col_ptr_end; X_col_ptr+=X_rows.step1(), x_col_ptr+=_x.step1() ) |
|
*x_col_ptr = (*X_col_ptr) / (*h_ptr); |
|
} |
|
} |
|
|
|
void |
|
homogeneousToEuclidean(const InputArray _X, OutputArray _x) |
|
{ |
|
// src |
|
const Mat X = _X.getMat(); |
|
|
|
// dst |
|
_x.create(X.rows-1, X.cols, X.type()); |
|
Mat x = _x.getMat(); |
|
|
|
// type |
|
if( X.depth() == CV_32F ) |
|
{ |
|
homogeneousToEuclidean<float>(X,x); |
|
} |
|
else |
|
{ |
|
homogeneousToEuclidean<double>(X,x); |
|
} |
|
} |
|
|
|
void |
|
euclideanToHomogeneous(const InputArray _x, OutputArray _X) |
|
{ |
|
const Mat x = _x.getMat(); |
|
const Mat last_row = Mat::ones(1, x.cols, x.type()); |
|
vconcat(x, last_row, _X); |
|
} |
|
|
|
template<typename T> |
|
void |
|
projectionFromKRt(const Mat_<T> &K, const Mat_<T> &R, const Mat_<T> &t, Mat_<T> P) |
|
{ |
|
hconcat( K*R, K*t, P ); |
|
} |
|
|
|
void |
|
projectionFromKRt(InputArray _K, InputArray _R, InputArray _t, OutputArray _P) |
|
{ |
|
const Mat K = _K.getMat(), R = _R.getMat(), t = _t.getMat(); |
|
const int depth = K.depth(); |
|
CV_Assert((K.cols == 3 && K.rows == 3) && (t.cols == 1 && t.rows == 3) && (K.size() == R.size())); |
|
CV_Assert((depth == CV_32F || depth == CV_64F) && depth == R.depth() && depth == t.depth()); |
|
|
|
_P.create(3, 4, depth); |
|
|
|
Mat P = _P.getMat(); |
|
|
|
// type |
|
if( depth == CV_32F ) |
|
{ |
|
projectionFromKRt<float>(K, R, t, P); |
|
} |
|
else |
|
{ |
|
projectionFromKRt<double>(K, R, t, P); |
|
} |
|
|
|
} |
|
|
|
template<typename T> |
|
void |
|
KRtFromProjection( const Mat_<T> &_P, Mat_<T> _K, Mat_<T> _R, Mat_<T> _t ) |
|
{ |
|
libmv::Mat34 P; |
|
libmv::Mat3 K, R; |
|
libmv::Vec3 t; |
|
|
|
cv2eigen( _P, P ); |
|
|
|
libmv::KRt_From_P( P, &K, &R, &t ); |
|
|
|
eigen2cv( K, _K ); |
|
eigen2cv( R, _R ); |
|
eigen2cv( t, _t ); |
|
} |
|
|
|
void |
|
KRtFromProjection( InputArray _P, OutputArray _K, OutputArray _R, OutputArray _t ) |
|
{ |
|
const Mat P = _P.getMat(); |
|
const int depth = P.depth(); |
|
CV_Assert((P.cols == 4 && P.rows == 3) && (depth == CV_32F || depth == CV_64F)); |
|
|
|
_K.create(3, 3, depth); |
|
_R.create(3, 3, depth); |
|
_t.create(3, 1, depth); |
|
|
|
Mat K = _K.getMat(), R = _R.getMat(), t = _t.getMat(); |
|
|
|
// type |
|
if( depth == CV_32F ) |
|
{ |
|
KRtFromProjection<float>(P, K, R, t); |
|
} |
|
else |
|
{ |
|
KRtFromProjection<double>(P, K, R, t); |
|
} |
|
} |
|
|
|
template<typename T> |
|
T |
|
depthValue( const Mat_<T> &_R, const Mat_<T> &_t, const Mat_<T> &_X ) |
|
{ |
|
Matx<T,3,3> R(_R); |
|
Vec<T,3> t(_t); |
|
|
|
if ( _X.rows == 3) |
|
{ |
|
Vec<T,3> X(_X); |
|
return (R*X)(2) + t(2); |
|
} |
|
else |
|
{ |
|
Vec<T,4> X(_X); |
|
Vec<T,3> Xe; |
|
homogeneousToEuclidean(X,Xe); |
|
return depthValue<T>( Mat(R), Mat(t), Mat(Xe) ); |
|
} |
|
} |
|
|
|
double |
|
depth( InputArray _R, InputArray _t, InputArray _X) |
|
{ |
|
const Mat R = _R.getMat(), t = _t.getMat(), X = _X.getMat(); |
|
const int depth = R.depth(); |
|
CV_Assert( R.rows == 3 && R.cols == 3 && t.rows == 3 && t.cols == 1 ); |
|
CV_Assert( (X.rows == 3 && X.cols == 1) || (X.rows == 4 && X.cols == 1) ); |
|
CV_Assert( depth == CV_32F || depth == CV_64F ); |
|
|
|
double depth_value = 0.0; |
|
|
|
if ( depth == CV_32F ) |
|
{ |
|
depth_value = static_cast<double>(depthValue<float>(R, t, X)); |
|
} |
|
else |
|
{ |
|
depth_value = depthValue<double>(R, t, X); |
|
} |
|
|
|
return depth_value; |
|
} |
|
|
|
} /* namespace sfm */ |
|
} /* namespace cv */
|
|
|