chessboard_stereo_camera_calib_internal.h

Go to the documentation of this file.

/* +------------------------------------------------------------------------+
   |                     Mobile Robot Programming Toolkit (MRPT)            |
   |                          https://www.mrpt.org/                         |
   |                                                                        |
   | Copyright (c) 2005-2019, Individual contributors, see AUTHORS file     |
   | See: https://www.mrpt.org/Authors - All rights reserved.               |
   | Released under BSD License. See: https://www.mrpt.org/License          |
   +------------------------------------------------------------------------+ */

#pragma once

#include <mrpt/math/geometry.h>
#include <mrpt/poses/CPose3D.h>
#include <mrpt/vision/chessboard_camera_calib.h>
#include <Eigen/Dense>

namespace mrpt::vision
{
// State of the Lev-Marq optimization:
struct lm_stat_t
{
    const TCalibrationStereoImageList& images;
    const std::vector<size_t>& valid_image_pair_indices;
    const std::vector<mrpt::math::TPoint3D>& obj_points;

    // State being optimized:
    //  N*left_cam_pose + right2left_pose + left_cam_params + right_cam_params
    std::vector<mrpt::poses::CPose3D>
        left_cam_poses;  // Poses of the origin of coordinates of the pattern
    // wrt the left camera
    mrpt::poses::CPose3D right2left_pose;
    /** [fx fy cx cy k1 k2 k3 t1 t2] */
    mrpt::math::CVectorFixedDouble<9> left_cam_params, right_cam_params;

    // Ctor
    lm_stat_t(
        const TCalibrationStereoImageList& _images,
        const std::vector<size_t>& _valid_image_pair_indices,
        const std::vector<mrpt::math::TPoint3D>& _obj_points)
        : images(_images),
          valid_image_pair_indices(_valid_image_pair_indices),
          obj_points(_obj_points)
    {
        left_cam_poses.assign(
            images.size(), mrpt::poses::CPose3D(0, 0, 1, 0, 0, 0));  // Initial
    }

    // Swap:
    void swap(lm_stat_t& o)
    {
        left_cam_poses.swap(o.left_cam_poses);
        std::swap(right2left_pose, o.right2left_pose);
        std::swap(left_cam_params, o.left_cam_params);
        std::swap(right_cam_params, o.right_cam_params);
    }
};

/** Data associated to *each observation* in the Lev-Marq. model */
struct TResidJacobElement
{
    /** [u_l v_l  u_r v_r]: left/right camera pixels */
    Eigen::Matrix<double, 4, 1> predicted_obs;
    /**  = predicted_obs - observations */
    Eigen::Matrix<double, 4, 1> residual;
    /** Jacobian. 4=the two predicted pixels; 30=Read below for the meaning of
     * these 30 variables */
    Eigen::Matrix<double, 4, 30> J;
};

using TResidualJacobianList = std::vector<std::vector<TResidJacobElement>>;

// Auxiliary functions for the Lev-Marq algorithm:
double recompute_errors_and_Jacobians(
    const lm_stat_t& lm_stat, TResidualJacobianList& res_jac,
    bool use_robust_kernel, double kernel_param);
void build_linear_system(
    const TResidualJacobianList& res_jac, const std::vector<size_t>& var_indxs,
    mrpt::math::CVectorDynamic<double>& minus_g, mrpt::math::CMatrixDouble& H);
void add_lm_increment(
    const mrpt::math::CVectorDynamic<double>& eps,
    const std::vector<size_t>& var_indxs, lm_stat_t& new_lm_stat);
}  // namespace mrpt::vision