总结
顶点
优化变量，pnp是位姿，后端里就是或者路标点和位姿。
需要自己定义优化变量维度，写初始化更新函数
初始化实现
边
观测值，用于优化，pnp里就是二维像素坐标，后端里就是。
需要自己写误差函数和雅克比矩阵函数
ch7 pnp 非线性优化g2o
pnp的误差函数
雅克比矩阵
g2o用于前端，即PNP
// A code block
var foo = 'bar';
/// vertex and edges used in g2o ba 优化顶点为6维位姿
class VertexPose : public g2o::BaseVertex<6, Sophus::SE3d> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
//设置初值
virtual void setToOriginImpl() override {
_estimate = Sophus::SE3d();
}
/// left multiplication on SE3 求解出的位姿更新量，
virtual void oplusImpl(const double *update) override {
Eigen::Matrix<double, 6, 1> update_eigen;
update_eigen << update[0], update[1], update[2], update[3], update[4], update[5];
_estimate = Sophus::SE3d::exp(update_eigen) * _estimate; //左乘更新量的指数映射，这里的李代数可以直接当做变换矩阵用
}
virtual bool read(istream &in) override {}
virtual bool write(ostream &out) const override {}
};
class EdgeProjection : public g2o::BaseUnaryEdge<2, Eigen::Vector2d, VertexPose> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
EdgeProjection(const Eigen::Vector3d &pos, const Eigen::Matrix3d &K) : _pos3d(pos), _K(K) {}
//计算pnp的误差，
virtual void computeError() override {
const VertexPose *v = static_cast<VertexPose *> (_vertices[0]); //待优化的位姿,顶点
Sophus::SE3d T = v->estimate();
Eigen::Vector3d pos_pixel = _K * (T * _pos3d);//转到归一化坐标
pos_pixel /= pos_pixel[2];//除以深度，获得像素坐标
_error = _measurement - pos_pixel.head<2>();//图2的观测值，减去图1的转到图2的预测值，
}
//雅克比矩阵
virtual void linearizeOplus() override {
const VertexPose *v = static_cast<VertexPose *> (_vertices[0]);
Sophus::SE3d T = v->estimate();
Eigen::Vector3d pos_cam = T * _pos3d; //转到图1的坐标系
double fx = _K(0, 0);
double fy = _K(1, 1);
double cx = _K(0, 2);
double cy = _K(1, 2);
double X = pos_cam[0];
double Y = pos_cam[1];
double Z = pos_cam[2];
double Z2 = Z * Z;
_jacobianOplusXi
<< -fx / Z, 0, fx * X / Z2, fx * X * Y / Z2, -fx - fx * X * X / Z2, fx * Y / Z,
0, -fy / Z, fy * Y / (Z * Z), fy + fy * Y * Y / Z2, -fy * X * Y / Z2, -fy * X / Z;
}
virtual bool read(istream &in) override {}
virtual bool write(ostream &out) const override {}
private:
Eigen::Vector3d _pos3d;
Eigen::Matrix3d _K;
};
void bundleAdjustmentG2O(
const VecVector3d &points_3d,
const VecVector2d &points_2d,
const Mat &K,
Sophus::SE3d &pose) {
// 构建图优化，先设定g2o
typedef g2o::BlockSolver<g2o::BlockSolverTraits<6, 3>> BlockSolverType; // pose is 6, landmark is 3
typedef g2o::LinearSolverDense<BlockSolverType::PoseMatrixType> LinearSolverType; // 线性求解器类型
// 梯度下降方法，可以从GN, LM, DogLeg 中选
auto solver = new g2o::OptimizationAlgorithmGaussNewton(
g2o::make_unique<BlockSolverType>(g2o::make_unique<LinearSolverType>()));
g2o::SparseOptimizer optimizer; // 图模型
optimizer.setAlgorithm(solver); // 设置求解器
optimizer.setVerbose(true); // 打开调试输出
// vertex
VertexPose *vertex_pose = new VertexPose(); // camera vertex_pose
vertex_pose->setId(0);//顶点id，只优化一个边，即位姿
vertex_pose->setEstimate(Sophus::SE3d()); //设置初始值
optimizer.addVertex(vertex_pose); //添加到优化器
// K
Eigen::Matrix3d K_eigen;
K_eigen <<
K.at<double>(0, 0), K.at<double>(0, 1), K.at<double>(0, 2),
K.at<double>(1, 0), K.at<double>(1, 1), K.at<double>(1, 2),
K.at<double>(2, 0), K.at<double>(2, 1), K.at<double>(2, 2);
// edges
int index = 1;
for (size_t i = 0; i < points_2d.size(); ++i) {
auto p2d = points_2d[i];
auto p3d = points_3d[i];
EdgeProjection *edge = new EdgeProjection(p3d, K_eigen);//边初始化，输入图2的世界坐标和内参
edge->setId(index);
edge->setVertex(0, vertex_pose);//设置边的序号和顶点
edge->setMeasurement(p2d); //图1的观测值
edge->setInformation(Eigen::Matrix2d::Identity());//设置信息矩阵
optimizer.addEdge(edge);
index++;
}
chrono::steady_clock::time_point t1 = chrono::steady_clock::now();
optimizer.setVerbose(true); 打开调试输出
optimizer.initializeOptimization();//设置优化信息
optimizer.optimize(10); //迭代10次
chrono::steady_clock::time_point t2 = chrono::steady_clock::now();
chrono::duration<double> time_used = chrono::duration_cast<chrono::duration<double>>(t2 - t1);
cout << "optimization costs time: " << time_used.count() << " seconds." << endl;
cout << "pose estimated by g2o =\n" << vertex_pose->estimate().matrix() << endl;
pose = vertex_pose->estimate();
}
g2o用于ICP。
3d-3d。优化变量同样是位姿，跟pnp的区别是少了内参，像素坐标，雅克比矩阵只有转换后的点对T求导，如下图。
实际中由于像素对应得深度可能没有，3D点不一定求得出来，所以pnp用的比较多。
/// vertex and edges used in g2o ba 顶点部分和pnp一样
class VertexPose : public g2o::BaseVertex<6, Sophus::SE3d> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
virtual void setToOriginImpl() override {
_estimate = Sophus::SE3d();
}
/// left multiplication on SE3
virtual void oplusImpl(const double *update) override {
Eigen::Matrix<double, 6, 1> update_eigen;
update_eigen << update[0], update[1], update[2], update[3], update[4], update[5];
_estimate = Sophus::SE3d::exp(update_eigen) * _estimate;
}
virtual bool read(istream &in) override {}
virtual bool write(ostream &out) const override {}
};
/// g2o edge
class EdgeProjectXYZRGBDPoseOnly : public g2o::BaseUnaryEdge<3, Eigen::Vector3d, VertexPose> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
EdgeProjectXYZRGBDPoseOnly(const Eigen::Vector3d &point) : _point(point) {}
//残差函数，图2的世界坐标乘变换矩阵，转到图1坐标系，图1减去图2的世界坐标系作为残差
virtual void computeError() override {
const VertexPose *pose = static_cast<const VertexPose *> ( _vertices[0] );
_error = _measurement - pose->estimate() * _point;
}
virtual void linearizeOplus() override {
VertexPose *pose = static_cast<VertexPose *>(_vertices[0]);
Sophus::SE3d T = pose->estimate();
Eigen::Vector3d xyz_trans = T * _point;//图2的世界坐标乘变换矩阵，转到图1坐标系
_jacobianOplusXi.block<3, 3>(0, 0) = -Eigen::Matrix3d::Identity();
_jacobianOplusXi.block<3, 3>(0, 3) = Sophus::SO3d::hat(xyz_trans);//Sophus::SO3d::hat将向量转为反对称矩阵，
}
bool read(istream &in) {}
bool write(ostream &out) const {}
protected:
Eigen::Vector3d _point;
};
g2o用于后端BAL
顶点
位姿和相机
/// 位姿加相机内参的顶点，9维，前三维为so3，接下去为t, f, k1, k2
class VertexPoseAndIntrinsics : public g2o::BaseVertex<9, PoseAndIntrinsics> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
VertexPoseAndIntrinsics() {}
virtual void setToOriginImpl() override {
_estimate = PoseAndIntrinsics(); //取出位姿和内参
}
virtual void oplusImpl(const double *update) override {
_estimate.rotation = SO3d::exp(Vector3d(update[0], update[1], update[2])) * _estimate.rotation;
_estimate.translation += Vector3d(update[3], update[4], update[5]);
_estimate.focal += update[6];
_estimate.k1 += update[7];
_estimate.k2 += update[8];
}
/// 根据估计值投影一个点，后面边计算误差时用
Vector2d project(const Vector3d &point) {
Vector3d pc = _estimate.rotation * point + _estimate.translation;
pc = -pc / pc[2]; //归一化坐标
double r2 = pc.squaredNorm(); //该像素点半径
double distortion = 1.0 + r2 * (_estimate.k1 + _estimate.k2 * r2);//去畸变
return Vector2d(_estimate.focal * distortion * pc[0],
_estimate.focal * distortion * pc[1]);
}
virtual bool read(istream &in) {}
virtual bool write(ostream &out) const {}
};
路标点
class VertexPoint : public g2o::BaseVertex<3, Vector3d> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
VertexPoint() {}
virtual void setToOriginImpl() override {
_estimate = Vector3d(0, 0, 0);
}
virtual void oplusImpl(const double *update) override {
_estimate += Vector3d(update[0], update[1], update[2]);
}
virtual bool read(istream &in) {}
virtual bool write(ostream &out) const {}
};
边
class EdgeProjection : //2为误差维度，Vector2d像素坐标，VertexPoseAndIntrinsics位姿和内参
public g2o::BaseBinaryEdge<2, Vector2d, VertexPoseAndIntrinsics, VertexPoint> {
public:
EIGEN_MAKE_ALIGNED_OPERATOR_NEW;
virtual void computeError() override {
auto v0 = (VertexPoseAndIntrinsics *) _vertices[0]; //位姿和内参
auto v1 = (VertexPoint *) _vertices[1]; //3d坐标
auto proj = v0->project(v1->estimate()); //求出2d像素坐标
_error = proj - _measurement;
}
// use numeric derivatives
virtual bool read(istream &in) {}
virtual bool write(ostream &out) const {}
};
main函数
void SolveBA(BALProblem &bal_problem) {
const int point_block_size = bal_problem.point_block_size();
const int camera_block_size = bal_problem.camera_block_size();
double *points = bal_problem.mutable_points();
double *cameras = bal_problem.mutable_cameras();
// pose dimension 9, landmark is 3
typedef g2o::BlockSolver<g2o::BlockSolverTraits<9, 3>> BlockSolverType;
typedef g2o::LinearSolverCSparse<BlockSolverType::PoseMatrixType> LinearSolverType;
// use LM
auto solver = new g2o::OptimizationAlgorithmLevenberg(
g2o::make_unique<BlockSolverType>(g2o::make_unique<LinearSolverType>()));
g2o::SparseOptimizer optimizer;
optimizer.setAlgorithm(solver);
optimizer.setVerbose(true);
/// build g2o problem
const double *observations = bal_problem.observations();
// vertex
vector<VertexPoseAndIntrinsics *> vertex_pose_intrinsics;
vector<VertexPoint *> vertex_points;
for (int i = 0; i < bal_problem.num_cameras(); ++i) {
VertexPoseAndIntrinsics *v = new VertexPoseAndIntrinsics();
double *camera = cameras + camera_block_size * i;
v->setId(i);
v->setEstimate(PoseAndIntrinsics(camera)); //pose和内参点的
optimizer.addVertex(v);
vertex_pose_intrinsics.push_back(v);
}
for (int i = 0; i < bal_problem.num_points(); ++i) {
VertexPoint *v = new VertexPoint();
double *point = points + point_block_size * i;
v->setId(i + bal_problem.num_cameras());
v->setEstimate(Vector3d(point[0], point[1], point[2]));//路标点顶点
// g2o在BA中需要手动设置待Marg的顶点，先固定路标点，求解出pose，再更新路标点
v->setMarginalized(true);
optimizer.addVertex(v);
vertex_points.push_back(v);
}
// edge
for (int i = 0; i < bal_problem.num_observations(); ++i) {
EdgeProjection *edge = new EdgeProjection;
edge->setVertex(0, vertex_pose_intrinsics[bal_problem.camera_index()[i]]); //pose和内参
edge->setVertex(1, vertex_points[bal_problem.point_index()[i]]);
edge->setMeasurement(Vector2d(observations[2 * i + 0], observations[2 * i + 1])); //2d像素坐标观测值
edge->setInformation(Matrix2d::Identity());
edge->setRobustKernel(new g2o::RobustKernelHuber());
optimizer.addEdge(edge);
}
optimizer.initializeOptimization();
optimizer.optimize(40);
// set to bal problem //设置新的pose和路标点
for (int i = 0; i < bal_problem.num_cameras(); ++i) {
double *camera = cameras + camera_block_size * i;
auto vertex = vertex_pose_intrinsics[i];
auto estimate = vertex->estimate();
estimate.set_to(camera);
}
for (int i = 0; i < bal_problem.num_points(); ++i) {
double *point = points + point_block_size * i;
auto vertex = vertex_points[i];
for (int k = 0; k < 3; ++k) point[k] = vertex->estimate()[k];
}
}
3513
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
