fangcywangermu
10/27/2019 - 9:49 AM
Optimizer类
/**
* This file is part of ORB-SLAM2.
*
* Copyright (C) 2014-2016 Ra煤l Mur-Artal <raulmur at unizar dot es> (University of Zaragoza)
* For more information see <https://github.com/raulmur/ORB_SLAM2>
*
* ORB-SLAM2 is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* ORB-SLAM2 is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with ORB-SLAM2. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef OPTIMIZER_H
#define OPTIMIZER_H
#include "Map.h"
#include "MapPoint.h"
#include "KeyFrame.h"
#include "LoopClosing.h"
#include "Frame.h"
#include "Thirdparty/g2o/g2o/types/types_seven_dof_expmap.h"
namespace ORB_SLAM2
{
class LoopClosing;
class Optimizer
{
public:
void static BundleAdjustment(const std::vector<KeyFrame*> &vpKF, const std::vector<MapPoint*> &vpMP,
int nIterations = 5, bool *pbStopFlag=NULL, const unsigned long nLoopKF=0,
const bool bRobust = true);
void static GlobalBundleAdjustemnt(Map* pMap, int nIterations=5, bool *pbStopFlag=NULL,
const unsigned long nLoopKF=0, const bool bRobust = true);
// 优化的是局部图
// 局部图中包括 一种地图点:共视关键帧的地图点
// 两种关键帧: 1 共视关键帧 2 可以看到共视关键帧地图点的其他非共视关键帧
// 优化的是共视关键帧位姿和共视关键帧的地图点坐标
void static LocalBundleAdjustment(KeyFrame* pKF, bool *pbStopFlag, Map *pMap);
// 优化位姿和地图点,并将地图点分为局外点和局内点
* 建立优化器
* 设置优化器的优化算法
* 添加位姿节点
* 添加地图点边
* 开始优化 执行四次迭代优化,每次优化都将所有点分为内点和外点,在下次优化时只根据上次结果的内点进行优化,计算出位姿后,对所有点重新分类内点/外点
* 将优化后的相机位姿添加到帧中 ,并将帧中的地图点分为内点和外点,返回内点数量
int static PoseOptimization(Frame* pFrame);
// if bFixScale is true, 6DoF optimization (stereo,rgbd), 7DoF otherwise (mono)
void static OptimizeEssentialGraph(Map* pMap, KeyFrame* pLoopKF, KeyFrame* pCurKF,
const LoopClosing::KeyFrameAndPose &NonCorrectedSim3,
const LoopClosing::KeyFrameAndPose &CorrectedSim3,
const map<KeyFrame *, set<KeyFrame *> > &LoopConnections,
const bool &bFixScale);
// if bFixScale is true, optimize SE3 (stereo,rgbd), Sim3 otherwise (mono)
static int OptimizeSim3(KeyFrame* pKF1, KeyFrame* pKF2, std::vector<MapPoint *> &vpMatches1,
g2o::Sim3 &g2oS12, const float th2, const bool bFixScale);
};
} //namespace ORB_SLAM
#endif // OPTIMIZER_H
/**
* This file is part of ORB-SLAM2.
*
* Copyright (C) 2014-2016 Raúl Mur-Artal <raulmur at unizar dot es> (University of Zaragoza)
* For more information see <https://github.com/raulmur/ORB_SLAM2>
*
* ORB-SLAM2 is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* ORB-SLAM2 is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with ORB-SLAM2. If not, see <http://www.gnu.org/licenses/>.
*/
#include "Optimizer.h"
#include "Thirdparty/g2o/g2o/core/block_solver.h"
#include "Thirdparty/g2o/g2o/core/optimization_algorithm_levenberg.h"
#include "Thirdparty/g2o/g2o/solvers/linear_solver_eigen.h"
#include "Thirdparty/g2o/g2o/types/types_six_dof_expmap.h"
#include "Thirdparty/g2o/g2o/core/robust_kernel_impl.h"
#include "Thirdparty/g2o/g2o/solvers/linear_solver_dense.h"
#include "Thirdparty/g2o/g2o/types/types_seven_dof_expmap.h"
#include<Eigen/StdVector>
#include "Converter.h"
#include<mutex>
namespace ORB_SLAM2
{
* 全局BA优化
* pMap(in): 整个地图
* nIterations(in): 迭代次数
* pbStopFlag(in): 停止标志量,表示是否该强制停止全局优化
* nLoopKF(in): 找到的回环帧的id
* bRobust(in): 是否需要鲁棒性内核
// pMap中所有的MapPoints和关键帧做bundle adjustment优化
// 这个全局BA优化在本程序中有两个地方使用：
// a.单目初始化：CreateInitialMapMonocular函数
// b.闭环优化：RunGlobalBundleAdjustment函数
void Optimizer::GlobalBundleAdjustemnt(Map* pMap, int nIterations, bool* pbStopFlag, const unsigned long nLoopKF, const bool bRobust)
{
vector<KeyFrame*> vpKFs = pMap->GetAllKeyFrames();
vector<MapPoint*> vpMP = pMap->GetAllMapPoints();
BundleAdjustment(vpKFs,vpMP,nIterations,pbStopFlag, nLoopKF, bRobust);
}
* 3D-2D 最小化重投影误差 e = (u,v) - project(Tcw*Pw)
* 1. Vertex: g2o::VertexSE3Expmap()，即当前帧的Tcw
* g2o::VertexSBAPointXYZ()，MapPoint的mWorldPos
* 2. Edge:
* - g2o::EdgeSE3ProjectXYZ()，BaseBinaryEdge
* + Vertex：待优化当前帧的Tcw
* + Vertex：待优化MapPoint的mWorldPos
* + measurement：MapPoint在当前帧中的二维位置(u,v)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* vpKFs(in): 整个地图中的关键帧
* vpMP(in): 整个地图中的地图点
* nIterations(in): 迭代次数
* pbStopFlag(in): 停止标志量,表示是否该强制停止全局优化
* nLoopKF(in): 找到的回环帧的id
* bRobust(in): 是否需要鲁棒性内核
void Optimizer::BundleAdjustment(const vector<KeyFrame *> &vpKFs, const vector<MapPoint *> &vpMP,
int nIterations, bool* pbStopFlag, const unsigned long nLoopKF, const bool bRobust)
{
vector<bool> vbNotIncludedMP;
vbNotIncludedMP.resize(vpMP.size());
// 步骤1：初始化g2o优化器
g2o::SparseOptimizer optimizer;
g2o::BlockSolver_6_3::LinearSolverType * linearSolver;
linearSolver = new g2o::LinearSolverEigen<g2o::BlockSolver_6_3::PoseMatrixType>();
g2o::BlockSolver_6_3 * solver_ptr = new g2o::BlockSolver_6_3(linearSolver);
g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg(solver_ptr);
optimizer.setAlgorithm(solver);
if(pbStopFlag)
optimizer.setForceStopFlag(pbStopFlag);
long unsigned int maxKFid = 0;
// 步骤2：向优化器添加顶点
// Set KeyFrame vertices
// 步骤2.1：向优化器添加关键帧位姿顶点
for(size_t i=0; i<vpKFs.size(); i++)
{
KeyFrame* pKF = vpKFs[i];
if(pKF->isBad())
continue;
g2o::VertexSE3Expmap * vSE3 = new g2o::VertexSE3Expmap();
vSE3->setEstimate(Converter::toSE3Quat(pKF->GetPose()));
vSE3->setId(pKF->mnId);
vSE3->setFixed(pKF->mnId==0);
optimizer.addVertex(vSE3);
if(pKF->mnId>maxKFid)
maxKFid=pKF->mnId;
}
const float thHuber2D = sqrt(5.99);
const float thHuber3D = sqrt(7.815);
// Set MapPoint vertices
// 步骤2.2：向优化器添加MapPoints顶点
for(size_t i=0; i<vpMP.size(); i++)
{
MapPoint* pMP = vpMP[i];
if(pMP->isBad())
continue;
g2o::VertexSBAPointXYZ* vPoint = new g2o::VertexSBAPointXYZ();
vPoint->setEstimate(Converter::toVector3d(pMP->GetWorldPos()));
const int id = pMP->mnId+maxKFid+1;
vPoint->setId(id);
vPoint->setMarginalized(true);
optimizer.addVertex(vPoint);
const map<KeyFrame*,size_t> observations = pMP->GetObservations();
int nEdges = 0;
//SET EDGES
// 步骤3：向优化器添加投影边边 两端分别是地图点三维坐标和观察到该三维点坐标的关键帧
for(map<KeyFrame*,size_t>::const_iterator mit=observations.begin(); mit!=observations.end(); mit++)
{
KeyFrame* pKF = mit->first;
if(pKF->isBad() || pKF->mnId>maxKFid)
continue;
nEdges++;
const cv::KeyPoint &kpUn = pKF->mvKeysUn[mit->second];
// 单目或RGBD相机
if(pKF->mvuRight[mit->second]<0)
{
Eigen::Matrix<double,2,1> obs;
obs << kpUn.pt.x, kpUn.pt.y;
g2o::EdgeSE3ProjectXYZ* e = new g2o::EdgeSE3ProjectXYZ();
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id)));
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pKF->mnId)));
e->setMeasurement(obs);
const float &invSigma2 = pKF->mvInvLevelSigma2[kpUn.octave];
e->setInformation(Eigen::Matrix2d::Identity()*invSigma2);
if(bRobust)
{
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber;
e->setRobustKernel(rk);
rk->setDelta(thHuber2D);
}
e->fx = pKF->fx;
e->fy = pKF->fy;
e->cx = pKF->cx;
e->cy = pKF->cy;
optimizer.addEdge(e);
}
else// 双目相机
{
Eigen::Matrix<double,3,1> obs;
const float kp_ur = pKF->mvuRight[mit->second];
obs << kpUn.pt.x, kpUn.pt.y, kp_ur;
g2o::EdgeStereoSE3ProjectXYZ* e = new g2o::EdgeStereoSE3ProjectXYZ();
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id)));
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pKF->mnId)));
e->setMeasurement(obs);
const float &invSigma2 = pKF->mvInvLevelSigma2[kpUn.octave];
Eigen::Matrix3d Info = Eigen::Matrix3d::Identity()*invSigma2;
e->setInformation(Info);
if(bRobust)
{
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber;
e->setRobustKernel(rk);
rk->setDelta(thHuber3D);
}
e->fx = pKF->fx;
e->fy = pKF->fy;
e->cx = pKF->cx;
e->cy = pKF->cy;
e->bf = pKF->mbf;
optimizer.addEdge(e);
}
}
if(nEdges==0)
{
optimizer.removeVertex(vPoint);
vbNotIncludedMP[i]=true;
}
else
{
vbNotIncludedMP[i]=false;
}
}
// Optimize!
// 步骤4：开始优化
optimizer.initializeOptimization();
optimizer.optimize(nIterations);
// Recover optimized data
// 步骤5：得到优化的结果
//Keyframes 根据优化结果恢复关键帧的位姿
for(size_t i=0; i<vpKFs.size(); i++)
{
KeyFrame* pKF = vpKFs[i];
if(pKF->isBad())
continue;
g2o::VertexSE3Expmap* vSE3 = static_cast<g2o::VertexSE3Expmap*>(optimizer.vertex(pKF->mnId));
g2o::SE3Quat SE3quat = vSE3->estimate();
if(nLoopKF==0)
{
pKF->SetPose(Converter::toCvMat(SE3quat));
}
else
{
pKF->mTcwGBA.create(4,4,CV_32F);
Converter::toCvMat(SE3quat).copyTo(pKF->mTcwGBA);
pKF->mnBAGlobalForKF = nLoopKF;
}
}
//Points 根据全局图优化恢复地图点的三维点坐标
for(size_t i=0; i<vpMP.size(); i++)
{
if(vbNotIncludedMP[i])
continue;
MapPoint* pMP = vpMP[i];
if(pMP->isBad())
continue;
g2o::VertexSBAPointXYZ* vPoint = static_cast<g2o::VertexSBAPointXYZ*>(optimizer.vertex(pMP->mnId+maxKFid+1));
if(nLoopKF==0)
{
pMP->SetWorldPos(Converter::toCvMat(vPoint->estimate()));
pMP->UpdateNormalAndDepth();
}
else
{
pMP->mPosGBA.create(3,1,CV_32F);
Converter::toCvMat(vPoint->estimate()).copyTo(pMP->mPosGBA);
pMP->mnBAGlobalForKF = nLoopKF;
}
}
}
// 优化位姿和地图点,并将地图点分为局外点和局内点
* 只优化Frame的Tcw，不优化MapPoints的坐标(只优化当前帧的pose，地图点固定)
// 该优化函数主要用于Tracking线程中：运动跟踪、参考帧跟踪、地图跟踪、重定位
* 建立优化器
* 设置优化器的优化算法
* 添加位姿节点
* 添加地图点边
* 开始优化 执行四次迭代优化,每次优化都将所有点分为内点和外点,在下次优化时只根据上次结果的内点进行优化,计算出位姿后,对所有点重新分类内点/外点
* 将优化后的相机位姿添加到帧中 ,并将帧中的地图点分为内点和外点,返回内点数量
* 3D-2D 最小化重投影误差 e = (u,v) - project(Tcw*Pw)
* 1. Vertex: g2o::VertexSE3Expmap()，即当前帧的Tcw
* 2. Edge:
* - g2o::EdgeSE3ProjectXYZOnlyPose()，BaseUnaryEdge
* + Vertex：待优化当前帧的Tcw
* + measurement：MapPoint在当前帧中的二维位置(u,v)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* - g2o::EdgeStereoSE3ProjectXYZOnlyPose()，BaseUnaryEdge
* + Vertex：待优化当前帧的Tcw
* + measurement：MapPoint在当前帧中的二维位置(ul,v,ur)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
// inliers数量
int Optimizer::PoseOptimization(Frame *pFrame)
{
// 步骤1：构造g2o优化器
g2o::SparseOptimizer optimizer;
g2o::BlockSolver_6_3::LinearSolverType * linearSolver;
linearSolver = new g2o::LinearSolverDense<g2o::BlockSolver_6_3::PoseMatrixType>();
g2o::BlockSolver_6_3 * solver_ptr = new g2o::BlockSolver_6_3(linearSolver);
g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg(solver_ptr);
optimizer.setAlgorithm(solver);
int nInitialCorrespondences=0;
// Set Frame vertex
// 步骤2：添加顶点：待优化当前帧的Tcw
g2o::VertexSE3Expmap * vSE3 = new g2o::VertexSE3Expmap();
vSE3->setEstimate(Converter::toSE3Quat(pFrame->mTcw));
vSE3->setId(0);
vSE3->setFixed(false);
optimizer.addVertex(vSE3);
// Set MapPoint vertices
const int N = pFrame->N;
vector<g2o::EdgeSE3ProjectXYZOnlyPose*> vpEdgesMono;
vector<size_t> vnIndexEdgeMono;
vpEdgesMono.reserve(N);
vnIndexEdgeMono.reserve(N);
vector<g2o::EdgeStereoSE3ProjectXYZOnlyPose*> vpEdgesStereo;
vector<size_t> vnIndexEdgeStereo;
vpEdgesStereo.reserve(N);
vnIndexEdgeStereo.reserve(N);
const float deltaMono = sqrt(5.991);
const float deltaStereo = sqrt(7.815);
// 步骤3：添加一元边：相机投影模型
{
unique_lock<mutex> lock(MapPoint::mGlobalMutex);
for(int i=0; i<N; i++)
{
MapPoint* pMP = pFrame->mvpMapPoints[i];
if(pMP)
{
// Monocular observation
// 单目情况, 也有可能在双目下, 当前帧的左兴趣点找不到匹配的右兴趣点
if(pFrame->mvuRight[i]<0)
{
nInitialCorrespondences++;
pFrame->mvbOutlier[i] = false;
Eigen::Matrix<double,2,1> obs;
const cv::KeyPoint &kpUn = pFrame->mvKeysUn[i];
obs << kpUn.pt.x, kpUn.pt.y;
g2o::EdgeSE3ProjectXYZOnlyPose* e = new g2o::EdgeSE3ProjectXYZOnlyPose();
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(0)));
e->setMeasurement(obs);
const float invSigma2 = pFrame->mvInvLevelSigma2[kpUn.octave];
e->setInformation(Eigen::Matrix2d::Identity()*invSigma2);
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber;
e->setRobustKernel(rk);
rk->setDelta(deltaMono);
e->fx = pFrame->fx;
e->fy = pFrame->fy;
e->cx = pFrame->cx;
e->cy = pFrame->cy;
cv::Mat Xw = pMP->GetWorldPos();
e->Xw[0] = Xw.at<float>(0);
e->Xw[1] = Xw.at<float>(1);
e->Xw[2] = Xw.at<float>(2);
optimizer.addEdge(e);
vpEdgesMono.push_back(e);
vnIndexEdgeMono.push_back(i);
}
else // Stereo observation 双目
{
nInitialCorrespondences++;
pFrame->mvbOutlier[i] = false;
//SET EDGE
Eigen::Matrix<double,3,1> obs; // 这里和单目不同
const cv::KeyPoint &kpUn = pFrame->mvKeysUn[i];
const float &kp_ur = pFrame->mvuRight[i];
obs << kpUn.pt.x, kpUn.pt.y, kp_ur; // 这里和单目不同
g2o::EdgeStereoSE3ProjectXYZOnlyPose* e = new g2o::EdgeStereoSE3ProjectXYZOnlyPose(); // 这里和单目不同
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(0)));
e->setMeasurement(obs);
const float invSigma2 = pFrame->mvInvLevelSigma2[kpUn.octave];
Eigen::Matrix3d Info = Eigen::Matrix3d::Identity()*invSigma2;
e->setInformation(Info);
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber;
e->setRobustKernel(rk);
rk->setDelta(deltaStereo);
e->fx = pFrame->fx;
e->fy = pFrame->fy;
e->cx = pFrame->cx;
e->cy = pFrame->cy;
e->bf = pFrame->mbf;
cv::Mat Xw = pMP->GetWorldPos();
e->Xw[0] = Xw.at<float>(0);
e->Xw[1] = Xw.at<float>(1);
e->Xw[2] = Xw.at<float>(2);
optimizer.addEdge(e);
vpEdgesStereo.push_back(e);
vnIndexEdgeStereo.push_back(i);
}
}
}
}
if(nInitialCorrespondences<3)
return 0;
// We perform 4 optimizations, after each optimization we classify observation as inlier/outlier
// At the next optimization, outliers are not included, but at the end they can be classified as inliers again.
// 步骤4：开始优化，总共优化四次，每次优化后，将观测分为outlier和inlier，outlier不参与下次优化,计算出位姿后,对所有点重新分类内点/外点
// 由于每次优化后是对所有的观测进行outlier和inlier判别，因此之前被判别为outlier有可能变成inlier，反之亦然
// 基于卡方检验计算出的阈值（假设测量有一个像素的偏差）
// 单目相机下允许的最大误差
const float chi2Mono[4]={5.991,5.991,5.991,5.991};
// 双目相机下允许的最大误差
const float chi2Stereo[4]={7.815,7.815,7.815, 7.815};
// 四次迭代，每次迭代的次数
const int its[4]={10,10,10,10};
// 外点数量
int nBad=0;
for(size_t it=0; it<4; it++)
{
vSE3->setEstimate(Converter::toSE3Quat(pFrame->mTcw));
// 对level为0的边进行优化
optimizer.initializeOptimization(0);
optimizer.optimize(its[it]);
nBad=0;
for(size_t i=0, iend=vpEdgesMono.size(); i<iend; i++)
{
g2o::EdgeSE3ProjectXYZOnlyPose* e = vpEdgesMono[i];
const size_t idx = vnIndexEdgeMono[i];
if(pFrame->mvbOutlier[idx])
{
e->computeError();
}
// 当前地图点的对于当前位姿模型下的误差
const float chi2 = e->chi2();
// 如果误差大于阈值 则证明是外点,该边的层数置为1
if(chi2>chi2Mono[it])
{
pFrame->mvbOutlier[idx]=true;
e->setLevel(1); // 设置为outlier
nBad++;
}
else // 否则误差小于阈值,则不是外点,边的层数置为0
{
pFrame->mvbOutlier[idx]=false;
e->setLevel(0); // 设置为inlier
}
if(it==2)
e->setRobustKernel(0); // 除了前两次优化需要RobustKernel以外, 其余的优化都不需要
}
for(size_t i=0, iend=vpEdgesStereo.size(); i<iend; i++)
{
g2o::EdgeStereoSE3ProjectXYZOnlyPose* e = vpEdgesStereo[i];
const size_t idx = vnIndexEdgeStereo[i];
if(pFrame->mvbOutlier[idx])
{
e->computeError();
}
const float chi2 = e->chi2();
if(chi2>chi2Stereo[it])
{
pFrame->mvbOutlier[idx]=true;
e->setLevel(1);
nBad++;
}
else
{
e->setLevel(0);
pFrame->mvbOutlier[idx]=false;
}
if(it==2)
e->setRobustKernel(0);
}
// 如果优化边数量小于10 也就是说内点数量太少 则退出优化
if(optimizer.edges().size()<10)
break;
}
// Recover optimized pose and return number of inliers
// 设置优化后的位姿和内点数量
g2o::VertexSE3Expmap* vSE3_recov = static_cast<g2o::VertexSE3Expmap*>(optimizer.vertex(0));
g2o::SE3Quat SE3quat_recov = vSE3_recov->estimate();
cv::Mat pose = Converter::toCvMat(SE3quat_recov);
pFrame->SetPose(pose);
return nInitialCorrespondences-nBad;
}
// 优化的是局部图
// 该优化函数用于LocalMapping线程的局部BA优化
// 局部图中包括 一种地图点: 共视关键帧的地图点
// 两种关键帧: 1 共视关键帧 2 可以看到共视关键帧地图点的其他非共视关键帧
// 优化的是共视关键帧位姿和共视关键帧的地图点坐标
* 1. Vertex:
* - g2o::VertexSE3Expmap()，LocalKeyFrames，即当前关键帧的位姿、与当前关键帧相连的关键帧的位姿
* - g2o::VertexSE3Expmap()，FixedCameras，即能观测到LocalMapPoints的关键帧（并且不属于LocalKeyFrames）的位姿，在优化中这些关键帧的位姿不变
* - g2o::VertexSBAPointXYZ()，LocalMapPoints，即LocalKeyFrames能观测到的所有MapPoints的位置
* 2. Edge:
* - g2o::EdgeSE3ProjectXYZ()，BaseBinaryEdge
* + Vertex：关键帧的Tcw，MapPoint的Pw
* + measurement：MapPoint在关键帧中的二维位置(u,v)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* - g2o::EdgeStereoSE3ProjectXYZ()，BaseBinaryEdge
* + Vertex：关键帧的Tcw，MapPoint的Pw
* + measurement：MapPoint在关键帧中的二维位置(ul,v,ur)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* pKF KeyFrame
* pbStopFlag 是否停止优化的标志
* pMap 在优化后，更新状态时需要用到Map的互斥量mMutexMapUpdate
void Optimizer::LocalBundleAdjustment(KeyFrame *pKF, bool* pbStopFlag, Map* pMap)
{
// Local KeyFrames: First Breath Search from Current Keyframe
list<KeyFrame*> lLocalKeyFrames;
// 步骤1：将当前关键帧加入lLocalKeyFrames
lLocalKeyFrames.push_back(pKF);
pKF->mnBALocalForKF = pKF->mnId;
// 步骤2：找到关键帧连接的关键帧（一级相连），加入lLocalKeyFrames中
const vector<KeyFrame*> vNeighKFs = pKF->GetVectorCovisibleKeyFrames();
for(int i=0, iend=vNeighKFs.size(); i<iend; i++)
{
KeyFrame* pKFi = vNeighKFs[i];
pKFi->mnBALocalForKF = pKF->mnId;
if(!pKFi->isBad())
lLocalKeyFrames.push_back(pKFi);
}
// Local MapPoints seen in Local KeyFrames
// 步骤3：遍历lLocalKeyFrames中关键帧，将它们观测的MapPoints加入到lLocalMapPoints
list<MapPoint*> lLocalMapPoints;
for(list<KeyFrame*>::iterator lit=lLocalKeyFrames.begin() , lend=lLocalKeyFrames.end(); lit!=lend; lit++)
{
vector<MapPoint*> vpMPs = (*lit)->GetMapPointMatches();
for(vector<MapPoint*>::iterator vit=vpMPs.begin(), vend=vpMPs.end(); vit!=vend; vit++)
{
MapPoint* pMP = *vit;
if(pMP)
if(!pMP->isBad())
if(pMP->mnBALocalForKF!=pKF->mnId)
{
lLocalMapPoints.push_back(pMP);
pMP->mnBALocalForKF=pKF->mnId; // 防止重复添加
}
}
}
// Fixed Keyframes. Keyframes that see Local MapPoints but that are not Local Keyframes
// 步骤4：得到能被局部MapPoints观测到，但不属于局部关键帧的关键帧，这些关键帧在局部BA优化时不优化(注意图优化的关键帧节点分为两部分vNeighKFs和lFixedCameras都有)
list<KeyFrame*> lFixedCameras;
for(list<MapPoint*>::iterator lit=lLocalMapPoints.begin(), lend=lLocalMapPoints.end(); lit!=lend; lit++)
{
map<KeyFrame*,size_t> observations = (*lit)->GetObservations();
for(map<KeyFrame*,size_t>::iterator mit=observations.begin(), mend=observations.end(); mit!=mend; mit++)
{
KeyFrame* pKFi = mit->first;
// pKFi->mnBALocalForKF!=pKF->mnId表示局部关键帧，此条件是为了将局部关键帧排除在外
// 其它的关键帧虽然能观测到，但不属于局部关键帧
if(pKFi->mnBALocalForKF!=pKF->mnId && pKFi->mnBAFixedForKF!=pKF->mnId)
{
pKFi->mnBAFixedForKF=pKF->mnId; // 防止重复添加
if(!pKFi->isBad())
lFixedCameras.push_back(pKFi);
}
}
}
// Setup optimizer
// 步骤5：构造g2o优化器
g2o::SparseOptimizer optimizer;
g2o::BlockSolver_6_3::LinearSolverType * linearSolver;
linearSolver = new g2o::LinearSolverEigen<g2o::BlockSolver_6_3::PoseMatrixType>();
g2o::BlockSolver_6_3 * solver_ptr = new g2o::BlockSolver_6_3(linearSolver);
g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg(solver_ptr);
optimizer.setAlgorithm(solver);
if(pbStopFlag)
optimizer.setForceStopFlag(pbStopFlag);
unsigned long maxKFid = 0;
// Set Local KeyFrame vertices
// 步骤6：添加顶点：Pose of Local KeyFrame,所有的局部关键帧
for(list<KeyFrame*>::iterator lit=lLocalKeyFrames.begin(), lend=lLocalKeyFrames.end(); lit!=lend; lit++)
{
KeyFrame* pKFi = *lit;
g2o::VertexSE3Expmap * vSE3 = new g2o::VertexSE3Expmap();
vSE3->setEstimate(Converter::toSE3Quat(pKFi->GetPose())); // Mat::cv转toSE3Quat 每一帧的姿态
vSE3->setId(pKFi->mnId);
vSE3->setFixed(pKFi->mnId==0); // 只有初始关键帧是固定位姿单位矩阵的 其余的关键帧位姿都可以根据图优化位姿进行调整
optimizer.addVertex(vSE3);
if(pKFi->mnId>maxKFid)
maxKFid=pKFi->mnId;
}
// Set Fixed KeyFrame vertices
// 步骤7：添加顶点：Pose of Fixed KeyFrame
// lFixedCameras:全部设为固定 (设置观察到局部地图点的非局部关键帧集作为图优化的帧节点)
for(list<KeyFrame*>::iterator lit=lFixedCameras.begin(), lend=lFixedCameras.end(); lit!=lend; lit++)
{
KeyFrame* pKFi = *lit;
g2o::VertexSE3Expmap * vSE3 = new g2o::VertexSE3Expmap();
vSE3->setEstimate(Converter::toSE3Quat(pKFi->GetPose()));
vSE3->setId(pKFi->mnId);
vSE3->setFixed(true);
optimizer.addVertex(vSE3);
if(pKFi->mnId>maxKFid)
maxKFid=pKFi->mnId;
}
// Set MapPoint vertices
// 步骤8: 设置地图点节点
const int nExpectedSize = (lLocalKeyFrames.size()+lFixedCameras.size())*lLocalMapPoints.size();
vector<g2o::EdgeSE3ProjectXYZ*> vpEdgesMono;
vpEdgesMono.reserve(nExpectedSize);
vector<KeyFrame*> vpEdgeKFMono;
vpEdgeKFMono.reserve(nExpectedSize);
vector<MapPoint*> vpMapPointEdgeMono;
vpMapPointEdgeMono.reserve(nExpectedSize);
vector<g2o::EdgeStereoSE3ProjectXYZ*> vpEdgesStereo;
vpEdgesStereo.reserve(nExpectedSize);
vector<KeyFrame*> vpEdgeKFStereo;
vpEdgeKFStereo.reserve(nExpectedSize);
vector<MapPoint*> vpMapPointEdgeStereo;
vpMapPointEdgeStereo.reserve(nExpectedSize);
const float thHuberMono = sqrt(5.991);
const float thHuberStereo = sqrt(7.815);
for(list<MapPoint*>::iterator lit=lLocalMapPoints.begin(), lend=lLocalMapPoints.end(); lit!=lend; lit++)
{
MapPoint* pMP = *lit;
g2o::VertexSBAPointXYZ* vPoint = new g2o::VertexSBAPointXYZ();
vPoint->setEstimate(Converter::toVector3d(pMP->GetWorldPos())); // 将OpenCVS中Mat类型的向量转化为Eigen中Matrix类型的变量
int id = pMP->mnId+maxKFid+1;
vPoint->setId(id);
vPoint->setMarginalized(true); // 边缘化
optimizer.addVertex(vPoint);
const map<KeyFrame*,size_t> observations = pMP->GetObservations();
//Set edges
// 步骤8：对每一对关联的MapPoint和KeyFrame构建边
for(map<KeyFrame*,size_t>::const_iterator mit=observations.begin(), mend=observations.end(); mit!=mend; mit++)
{
KeyFrame* pKFi = mit->first;
if(!pKFi->isBad())
{
const cv::KeyPoint &kpUn = pKFi->mvKeysUn[mit->second];
// Monocular observation 单目
if(pKFi->mvuRight[mit->second]<0)
{
Eigen::Matrix<double,2,1> obs; //特征点坐标
obs << kpUn.pt.x, kpUn.pt.y;
g2o::EdgeSE3ProjectXYZ* e = new g2o::EdgeSE3ProjectXYZ();
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id))); //0: mappoints
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pKFi->mnId))); //1: 关键帧位姿
e->setMeasurement(obs); //观测量:像素坐标(kpUn.pt.x, kpUn.pt.y)
const float &invSigma2 = pKFi->mvInvLevelSigma2[kpUn.octave];
e->setInformation(Eigen::Matrix2d::Identity()*invSigma2);
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber; //Huber核函数
e->setRobustKernel(rk);
rk->setDelta(thHuberMono); //误差的界限err<2.44
e->fx = pKFi->fx;
e->fy = pKFi->fy;
e->cx = pKFi->cx;
e->cy = pKFi->cy;
optimizer.addEdge(e);
vpEdgesMono.push_back(e);
vpEdgeKFMono.push_back(pKFi);
vpMapPointEdgeMono.push_back(pMP);
}
else // Stereo observation 双目
{
Eigen::Matrix<double,3,1> obs;
const float kp_ur = pKFi->mvuRight[mit->second];
obs << kpUn.pt.x, kpUn.pt.y, kp_ur;
g2o::EdgeStereoSE3ProjectXYZ* e = new g2o::EdgeStereoSE3ProjectXYZ();
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id)));
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pKFi->mnId)));
e->setMeasurement(obs);
const float &invSigma2 = pKFi->mvInvLevelSigma2[kpUn.octave];
Eigen::Matrix3d Info = Eigen::Matrix3d::Identity()*invSigma2;
e->setInformation(Info);
g2o::RobustKernelHuber* rk = new g2o::RobustKernelHuber;
e->setRobustKernel(rk);
rk->setDelta(thHuberStereo);
e->fx = pKFi->fx;
e->fy = pKFi->fy;
e->cx = pKFi->cx;
e->cy = pKFi->cy;
e->bf = pKFi->mbf;
optimizer.addEdge(e);
vpEdgesStereo.push_back(e);
vpEdgeKFStereo.push_back(pKFi);
vpMapPointEdgeStereo.push_back(pMP);
}
}
}
}
if(pbStopFlag)
if(*pbStopFlag)
return;
// 步骤9: 优化一次,筛选内点后再优化一次进一步筛选内点,得到最终的结果
optimizer.initializeOptimization();
optimizer.optimize(5);
bool bDoMore= true;
if(pbStopFlag)
if(*pbStopFlag)
bDoMore = false;
if(bDoMore)
{
// Check inlier observations
// 步骤10：检测outlier，并设置下次不优化
for(size_t i=0, iend=vpEdgesMono.size(); i<iend;i++)
{
g2o::EdgeSE3ProjectXYZ* e = vpEdgesMono[i];
MapPoint* pMP = vpMapPointEdgeMono[i];
if(pMP->isBad())
continue;
// 基于卡方检验计算出的阈值（假设测量有一个像素的偏差）
// return _error.dot(information()*_error);
if(e->chi2()>5.991 || !e->isDepthPositive())
{
e->setLevel(1); // 不优化
}
e->setRobustKernel(0); // 不使用核函数
}
for(size_t i=0, iend=vpEdgesStereo.size(); i<iend;i++)
{
g2o::EdgeStereoSE3ProjectXYZ* e = vpEdgesStereo[i];
MapPoint* pMP = vpMapPointEdgeStereo[i];
if(pMP->isBad())
continue;
if(e->chi2()>7.815 || !e->isDepthPositive())
{
e->setLevel(1);
}
e->setRobustKernel(0);
}
// Optimize again without the outliers
// 步骤11：排除误差较大的outlier后再次优化
optimizer.initializeOptimization(0);
optimizer.optimize(10);
}
vector<pair<KeyFrame*,MapPoint*> > vToErase;
vToErase.reserve(vpEdgesMono.size()+vpEdgesStereo.size());
// Check inlier observations
// 步骤12：在优化后重新计算误差，剔除连接误差比较大的关键帧和MapPoint
for(size_t i=0, iend=vpEdgesMono.size(); i<iend;i++)
{
g2o::EdgeSE3ProjectXYZ* e = vpEdgesMono[i];
MapPoint* pMP = vpMapPointEdgeMono[i];
if(pMP->isBad())
continue;
// 基于卡方检验计算出的阈值（假设测量有一个像素的偏差
if(e->chi2()>5.991 || !e->isDepthPositive())
{
KeyFrame* pKFi = vpEdgeKFMono[i];
vToErase.push_back(make_pair(pKFi,pMP));
}
}
for(size_t i=0, iend=vpEdgesStereo.size(); i<iend;i++)
{
g2o::EdgeStereoSE3ProjectXYZ* e = vpEdgesStereo[i];
MapPoint* pMP = vpMapPointEdgeStereo[i];
if(pMP->isBad())
continue;
if(e->chi2()>7.815 || !e->isDepthPositive())
{
KeyFrame* pKFi = vpEdgeKFStereo[i];
vToErase.push_back(make_pair(pKFi,pMP));
}
}
// Get Map Mutex
unique_lock<mutex> lock(pMap->mMutexMapUpdate);
// 连接偏差比较大，在关键帧中剔除对该MapPoint的观测
// 连接偏差比较大，在MapPoint中剔除对该关键帧的观测
if(!vToErase.empty())
{
for(size_t i=0;i<vToErase.size();i++)
{
KeyFrame* pKFi = vToErase[i].first;
MapPoint* pMPi = vToErase[i].second;
pKFi->EraseMapPointMatch(pMPi);
pMPi->EraseObservation(pKFi);
}
}
// Recover optimized data
// 步骤13：优化后更新关键帧位姿以及MapPoints的位置、平均观测方向等属性
//Keyframes
for(list<KeyFrame*>::iterator lit=lLocalKeyFrames.begin(), lend=lLocalKeyFrames.end(); lit!=lend; lit++)
{
KeyFrame* pKF = *lit;
g2o::VertexSE3Expmap* vSE3 = static_cast<g2o::VertexSE3Expmap*>(optimizer.vertex(pKF->mnId));
g2o::SE3Quat SE3quat = vSE3->estimate();
pKF->SetPose(Converter::toCvMat(SE3quat));
}
//Points
for(list<MapPoint*>::iterator lit=lLocalMapPoints.begin(), lend=lLocalMapPoints.end(); lit!=lend; lit++)
{
MapPoint* pMP = *lit;
g2o::VertexSBAPointXYZ* vPoint = static_cast<g2o::VertexSBAPointXYZ*>(optimizer.vertex(pMP->mnId+maxKFid+1));
pMP->SetWorldPos(Converter::toCvMat(vPoint->estimate()));
pMP->UpdateNormalAndDepth();
}
}
pMap 全局地图
pLoopKF 闭环匹配上的关键帧
pCurKF 当前关键帧
NonCorrectedSim3 未经过Sim3传播调整过的关键帧位姿
CorrectedSim3 经过Sim3传播调整过的关键帧位姿
LoopConnections 因闭环时MapPoints调整而新生成的边
闭环检测后，EssentialGraph优化
* 1. Vertex:
* - g2o::VertexSim3Expmap，Essential graph中关键帧的位姿
* 2. Edge:
* - g2o::EdgeSim3()，BaseBinaryEdge
* + Vertex：关键帧的Tcw，MapPoint的Pw
* + measurement：经过CorrectLoop函数步骤2，Sim3传播校正后的位姿
* + InfoMatrix: 单位矩阵
void Optimizer::OptimizeEssentialGraph(Map* pMap, KeyFrame* pLoopKF, KeyFrame* pCurKF,
const LoopClosing::KeyFrameAndPose &NonCorrectedSim3,
const LoopClosing::KeyFrameAndPose &CorrectedSim3,
const map<KeyFrame *, set<KeyFrame *> > &LoopConnections, const bool &bFixScale)
{
// Setup optimizer
// 步骤1：构造优化器
g2o::SparseOptimizer optimizer;
optimizer.setVerbose(false);
// 指定线性方程求解器使用Eigen的块求解器
// BlockSolver_7_3: solver fo BA with scale
g2o::BlockSolver_7_3::LinearSolverType * linearSolver =
new g2o::LinearSolverEigen<g2o::BlockSolver_7_3::PoseMatrixType>();
// 构造线性求解器
g2o::BlockSolver_7_3 * solver_ptr= new g2o::BlockSolver_7_3(linearSolver);
// 使用LM算法进行非线性迭代
g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg(solver_ptr);
solver->setUserLambdaInit(1e-16);
optimizer.setAlgorithm(solver);
const vector<KeyFrame*> vpKFs = pMap->GetAllKeyFrames();
const vector<MapPoint*> vpMPs = pMap->GetAllMapPoints();
//最后一帧关键帧
const unsigned int nMaxKFid = pMap->GetMaxKFid();
// 仅经过Sim3传播调整，未经过优化的keyframe的pose
vector<g2o::Sim3,Eigen::aligned_allocator<g2o::Sim3> > vScw(nMaxKFid+1);
// 经过Sim3传播调整，经过优化的keyframe的pose
vector<g2o::Sim3,Eigen::aligned_allocator<g2o::Sim3> > vCorrectedSwc(nMaxKFid+1);
vector<g2o::VertexSim3Expmap*> vpVertices(nMaxKFid+1);
const int minFeat = 100;
// Set KeyFrame vertices
// 步骤2：将地图中所有keyframe的pose作为顶点添加到优化器
// 地图中的关键帧的位姿全部优先以修正过的位姿为顶点
// 尽可能使用经过Sim3调整的位姿
for(size_t i=0, iend=vpKFs.size(); i<iend;i++)
{
KeyFrame* pKF = vpKFs[i];
if(pKF->isBad())
continue;
g2o::VertexSim3Expmap* VSim3 = new g2o::VertexSim3Expmap();
const int nIDi = pKF->mnId;
LoopClosing::KeyFrameAndPose::const_iterator it = CorrectedSim3.find(pKF);
// 如果该关键帧在闭环时通过Sim3传播调整过，用校正后的位姿
if(it!=CorrectedSim3.end())
{
vScw[nIDi] = it->second;
VSim3->setEstimate(it->second);
}
else // 如果该关键帧在闭环时没有通过Sim3传播调整过，用自身的位姿
{
Eigen::Matrix<double,3,3> Rcw = Converter::toMatrix3d(pKF->GetRotation());
Eigen::Matrix<double,3,1> tcw = Converter::toVector3d(pKF->GetTranslation());
g2o::Sim3 Siw(Rcw,tcw,1.0);
vScw[nIDi] = Siw;
VSim3->setEstimate(Siw);
}
// 如果是回环的两端的关键帧,则不进行优化
if(pKF==pLoopKF)
VSim3->setFixed(true);
VSim3->setId(nIDi);
VSim3->setMarginalized(false);
VSim3->_fix_scale = bFixScale;
optimizer.addVertex(VSim3);
// 优化前的pose顶点，后面代码中没有使用
vpVertices[nIDi]=VSim3;
}
set<pair<long unsigned int,long unsigned int> > sInsertedEdges;
const Eigen::Matrix<double,7,7> matLambda = Eigen::Matrix<double,7,7>::Identity();
// Set Loop edges
// 设置回环边 边的两端分别是回环的两个端点,回环的初始端点是一个关键帧集合,末端点与该关键帧集合形成若干的边
// 步骤3：添加边：LoopConnections是闭环时因为MapPoints调整而出现的新关键帧连接关系（不是当前帧与闭环匹配帧之间的连接关系
for(map<KeyFrame *, set<KeyFrame *> >::const_iterator mit = LoopConnections.begin(), mend=LoopConnections.end(); mit!=mend; mit++)
{
KeyFrame* pKF = mit->first;
const long unsigned int nIDi = pKF->mnId;
const set<KeyFrame*> &spConnections = mit->second;
const g2o::Sim3 Siw = vScw[nIDi];
const g2o::Sim3 Swi = Siw.inverse();
for(set<KeyFrame*>::const_iterator sit=spConnections.begin(), send=spConnections.end(); sit!=send; sit++)
{
const long unsigned int nIDj = (*sit)->mnId;
if((nIDi!=pCurKF->mnId || nIDj!=pLoopKF->mnId) && pKF->GetWeight(*sit)<minFeat)
continue;
const g2o::Sim3 Sjw = vScw[nIDj];
// 得到两个pose间的Sim3变换
const g2o::Sim3 Sji = Sjw * Swi;
g2o::EdgeSim3* e = new g2o::EdgeSim3();
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDj)));
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDi)));
// 根据两个Pose顶点的位姿算出相对位姿作为边，那还存在误差？优化有用？因为闭环MapPoints调整新形成的边不优化？（wubo???）
e->setMeasurement(Sji);
e->information() = matLambda;
optimizer.addEdge(e);
sInsertedEdges.insert(make_pair(min(nIDi,nIDj),max(nIDi,nIDj)));
}
}
// Set normal edges
// 步骤4：添加跟踪时形成的边、闭环匹配成功形成的边
for(size_t i=0, iend=vpKFs.size(); i<iend; i++)
{
KeyFrame* pKF = vpKFs[i];
const int nIDi = pKF->mnId;
g2o::Sim3 Swi;
LoopClosing::KeyFrameAndPose::const_iterator iti = NonCorrectedSim3.find(pKF);
// 尽可能得到未经过Sim3传播调整的位姿
if(iti!=NonCorrectedSim3.end())
Swi = (iti->second).inverse();
else // 矫正后的位姿
Swi = vScw[nIDi].inverse();
KeyFrame* pParentKF = pKF->GetParent();
// Spanning tree edge
// 步骤4.1：只添加扩展树的边（有父关键帧）
// 第一种边: 边两端为父关键帧位姿(未矫正的相似变换矩阵)和当前关键帧位姿(未矫正的相似变换矩阵)
// 顶点(矫正后) ------普通关键帧与父关键帧的位姿变换关系(未矫正前)-----顶点(矫正后)
if(pParentKF)
{
int nIDj = pParentKF->mnId;
g2o::Sim3 Sjw;
LoopClosing::KeyFrameAndPose::const_iterator itj = NonCorrectedSim3.find(pParentKF);
// 尽可能得到未经过Sim3传播调整的位姿
if(itj!=NonCorrectedSim3.end())
Sjw = itj->second;
else
Sjw = vScw[nIDj];
// Sji:当前帧到父关键帧的变换
g2o::Sim3 Sji = Sjw * Swi;
g2o::EdgeSim3* e = new g2o::EdgeSim3();
e->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDj)));
e->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDi)));
e->setMeasurement(Sji);
e->information() = matLambda;
optimizer.addEdge(e);
}
// Loop edges
// 步骤4.2：添加在CorrectLoop函数中AddLoopEdge函数添加的闭环连接边（当前帧与闭环匹配帧之间的连接关系）
// 使用经过Sim3调整前关键帧之间的相对关系作为边
// 第二种边: 回环边,边的两边分别是当前关键帧和其回环关键帧的位姿(未矫正的相似变换矩阵)
// 顶点(矫正后) ------普通关键帧与父关键帧的位姿变换关系(未矫正前)-----顶点(矫正后)
const set<KeyFrame*> sLoopEdges = pKF->GetLoopEdges();
for(set<KeyFrame*>::const_iterator sit=sLoopEdges.begin(), send=sLoopEdges.end(); sit!=send; sit++)
{
KeyFrame* pLKF = *sit;
if(pLKF->mnId<pKF->mnId)
{
g2o::Sim3 Slw;
LoopClosing::KeyFrameAndPose::const_iterator itl = NonCorrectedSim3.find(pLKF);
// 尽可能得到未经过Sim3传播调整的位姿
if(itl!=NonCorrectedSim3.end())
Slw = itl->second;
else
Slw = vScw[pLKF->mnId];
g2o::Sim3 Sli = Slw * Swi;
g2o::EdgeSim3* el = new g2o::EdgeSim3();
el->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pLKF->mnId)));
el->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDi)));
// 根据两个Pose顶点的位姿算出相对位姿作为边，那还存在误差？优化有用？（wubo???）
el->setMeasurement(Sli);
el->information() = matLambda;
optimizer.addEdge(el);
}
}
// Covisibility graph edges
// 步骤4.3：最有很好共视关系的关键帧也作为边进行优化
// 使用经过Sim3调整前关键帧之间的相对关系作为边
// 第三种边:共视图边 边的两端分别是当前关键帧位姿和共视关键帧的位姿(未矫正的相似变换矩阵)
const vector<KeyFrame*> vpConnectedKFs = pKF->GetCovisiblesByWeight(minFeat);
for(vector<KeyFrame*>::const_iterator vit=vpConnectedKFs.begin(); vit!=vpConnectedKFs.end(); vit++)
{
KeyFrame* pKFn = *vit;
if(pKFn && pKFn!=pParentKF && !pKF->hasChild(pKFn) && !sLoopEdges.count(pKFn))
{
if(!pKFn->isBad() && pKFn->mnId<pKF->mnId)
{
if(sInsertedEdges.count(make_pair(min(pKF->mnId,pKFn->mnId),max(pKF->mnId,pKFn->mnId))))
continue;
g2o::Sim3 Snw;
LoopClosing::KeyFrameAndPose::const_iterator itn = NonCorrectedSim3.find(pKFn);
// 尽可能得到未经过Sim3传播调整的位姿
if(itn!=NonCorrectedSim3.end())
Snw = itn->second;
else
Snw = vScw[pKFn->mnId];
g2o::Sim3 Sni = Snw * Swi;
g2o::EdgeSim3* en = new g2o::EdgeSim3();
en->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(pKFn->mnId)));
en->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(nIDi)));
en->setMeasurement(Sni);
en->information() = matLambda;
optimizer.addEdge(en);
}
}
}
}
// Optimize!
// 步骤5：开始g2o优化
optimizer.initializeOptimization();
optimizer.optimize(20);
unique_lock<mutex> lock(pMap->mMutexMapUpdate);
// SE3 Pose Recovering. Sim3:[sR t;0 1] -> SE3:[R t/s;0 1]
// 步骤6：设定优化后的位姿（位姿恢复 从相似变换矩阵变换到欧式变换矩阵）
for(size_t i=0;i<vpKFs.size();i++)
{
KeyFrame* pKFi = vpKFs[i];
const int nIDi = pKFi->mnId;
g2o::VertexSim3Expmap* VSim3 = static_cast<g2o::VertexSim3Expmap*>(optimizer.vertex(nIDi));
g2o::Sim3 CorrectedSiw = VSim3->estimate();
vCorrectedSwc[nIDi]=CorrectedSiw.inverse();
Eigen::Matrix3d eigR = CorrectedSiw.rotation().toRotationMatrix();
Eigen::Vector3d eigt = CorrectedSiw.translation();
double s = CorrectedSiw.scale();
eigt *=(1./s); //[R t/s;0 1]
cv::Mat Tiw = Converter::toCvSE3(eigR,eigt);
pKFi->SetPose(Tiw);
}
// Correct points. Transform to "non-optimized" reference keyframe pose and transform back with optimized pose
// 步骤7：步骤5和步骤6优化得到关键帧的位姿后，MapPoints根据参考帧优化前后的相对关系调整自己的位置
for(size_t i=0, iend=vpMPs.size(); i<iend; i++)
{
MapPoint* pMP = vpMPs[i];
if(pMP->isBad())
continue;
int nIDr;
if(pMP->mnCorrectedByKF==pCurKF->mnId)
{
nIDr = pMP->mnCorrectedReference;
}
else
{
KeyFrame* pRefKF = pMP->GetReferenceKeyFrame();
nIDr = pRefKF->mnId;
}
// 得到MapPoint参考关键帧步骤5优化前的位姿
g2o::Sim3 Srw = vScw[nIDr];
// 得到MapPoint参考关键帧优化后的位姿
g2o::Sim3 correctedSwr = vCorrectedSwc[nIDr];
cv::Mat P3Dw = pMP->GetWorldPos();
Eigen::Matrix<double,3,1> eigP3Dw = Converter::toVector3d(P3Dw);
Eigen::Matrix<double,3,1> eigCorrectedP3Dw = correctedSwr.map(Srw.map(eigP3Dw));
cv::Mat cvCorrectedP3Dw = Converter::toCvMat(eigCorrectedP3Dw);
pMP->SetWorldPos(cvCorrectedP3Dw);
pMP->UpdateNormalAndDepth();
}
}
// 优化形成闭环时进行Sim3优化
* 对RANSAC算出当前两帧的Sim3变换结果,通过BA优化一下
* 顶点:
* 1. 两个关键帧的位姿
* 2. 两个关键帧共有的MapPoints
*
* 边: 有2种:
* 1. kf1的地图点投影(Sim3)到kf2上;
* 2. kf2的地图点投影(Sim3)到kf1上
* 1. Vertex:
* - g2o::VertexSim3Expmap()，两个关键帧的位姿
* - g2o::VertexSBAPointXYZ()，两个关键帧共有的MapPoints
* 2. Edge:
* - g2o::EdgeSim3ProjectXYZ()，BaseBinaryEdge
* + Vertex：关键帧的Sim3，MapPoint的Pw
* + measurement：MapPoint在关键帧中的二维位置(u,v)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* - g2o::EdgeInverseSim3ProjectXYZ()，BaseBinaryEdge
* + Vertex：关键帧的Sim3，MapPoint的Pw
* + measurement：MapPoint在关键帧中的二维位置(u,v)
* + InfoMatrix: invSigma2(与特征点所在的尺度有关)
* pKF1 KeyFrame
* pKF2 KeyFrame
* vpMatches1 两个关键帧的匹配关系
* g2oS12 两个关键帧间的Sim3变换
* th2 核函数阈值
* bFixScale 是否优化尺度，单目进行尺度优化，双目不进行尺度优化
int Optimizer::OptimizeSim3(KeyFrame *pKF1, KeyFrame *pKF2, vector<MapPoint *> &vpMatches1, g2o::Sim3 &g2oS12, const float th2, const bool bFixScale)
{
// 步骤1：初始化g2o优化器
// 先构造求解器
g2o::SparseOptimizer optimizer;
// 构造线性方程求解器，Hx = -b的求解器
g2o::BlockSolverX::LinearSolverType * linearSolver;
// 使用dense的求解器，（常见非dense求解器有cholmod线性求解器和shur补线性求解器）
linearSolver = new g2o::LinearSolverDense<g2o::BlockSolverX::PoseMatrixType>();
g2o::BlockSolverX * solver_ptr = new g2o::BlockSolverX(linearSolver);
// 使用L-M迭代
g2o::OptimizationAlgorithmLevenberg* solver = new g2o::OptimizationAlgorithmLevenberg(solver_ptr);
optimizer.setAlgorithm(solver);
// Calibration
const cv::Mat &K1 = pKF1->mK;
const cv::Mat &K2 = pKF2->mK;
// Camera poses
const cv::Mat R1w = pKF1->GetRotation();
const cv::Mat t1w = pKF1->GetTranslation();
const cv::Mat R2w = pKF2->GetRotation();
const cv::Mat t2w = pKF2->GetTranslation();
// Set Sim3 vertex
// 步骤2.1 添加Sim3顶点
g2o::VertexSim3Expmap * vSim3 = new g2o::VertexSim3Expmap();
vSim3->_fix_scale=bFixScale; //是否优化尺度
vSim3->setEstimate(g2oS12); //优化sim3位姿
vSim3->setId(0);
vSim3->setFixed(false); //优化Sim3顶点
vSim3->_principle_point1[0] = K1.at<float>(0,2); // 光心横坐标cx
vSim3->_principle_point1[1] = K1.at<float>(1,2); // 光心纵坐标cy
vSim3->_focal_length1[0] = K1.at<float>(0,0); // 焦距 fx
vSim3->_focal_length1[1] = K1.at<float>(1,1); // 焦距 fy
vSim3->_principle_point2[0] = K2.at<float>(0,2);
vSim3->_principle_point2[1] = K2.at<float>(1,2);
vSim3->_focal_length2[0] = K2.at<float>(0,0);
vSim3->_focal_length2[1] = K2.at<float>(1,1);
optimizer.addVertex(vSim3);
// Set MapPoint vertices
const int N = vpMatches1.size();
const vector<MapPoint*> vpMapPoints1 = pKF1->GetMapPointMatches();
//pKF2对应的MapPoints到pKF1的投影
vector<g2o::EdgeSim3ProjectXYZ*> vpEdges12;
//pKF1对应的MapPoints到pKF2的投影
vector<g2o::EdgeInverseSim3ProjectXYZ*> vpEdges21;
vector<size_t> vnIndexEdge;
vnIndexEdge.reserve(2*N);
vpEdges12.reserve(2*N);
vpEdges21.reserve(2*N);
const float deltaHuber = sqrt(th2);
int nCorrespondences = 0;
for(int i=0; i<N; i++)
{
if(!vpMatches1[i])
continue;
// pMP1和pMP2是匹配的MapPoints
MapPoint* pMP1 = vpMapPoints1[i];
MapPoint* pMP2 = vpMatches1[i];
const int id1 = 2*i+1;
const int id2 = 2*(i+1);
const int i2 = pMP2->GetIndexInKeyFrame(pKF2);
if(pMP1 && pMP2)
{
if(!pMP1->isBad() && !pMP2->isBad() && i2>=0)
{
// 步骤2.2 添加PointXYZ顶点
g2o::VertexSBAPointXYZ* vPoint1 = new g2o::VertexSBAPointXYZ();
cv::Mat P3D1w = pMP1->GetWorldPos();
cv::Mat P3D1c = R1w*P3D1w + t1w;
vPoint1->setEstimate(Converter::toVector3d(P3D1c));
vPoint1->setId(id1);
vPoint1->setFixed(true); //固定,不优化
optimizer.addVertex(vPoint1);
g2o::VertexSBAPointXYZ* vPoint2 = new g2o::VertexSBAPointXYZ();
cv::Mat P3D2w = pMP2->GetWorldPos();
cv::Mat P3D2c = R2w*P3D2w + t2w;
vPoint2->setEstimate(Converter::toVector3d(P3D2c));
vPoint2->setId(id2);
vPoint2->setFixed(true); //固定,不优化
optimizer.addVertex(vPoint2);
}
else
continue;
}
else
continue;
nCorrespondences++;
// Set edge x1 = S12*X2
Eigen::Matrix<double,2,1> obs1;
const cv::KeyPoint &kpUn1 = pKF1->mvKeysUn[i];
obs1 << kpUn1.pt.x, kpUn1.pt.y;
// 步骤2.3 添加两个顶点（3D点）到相机投影的边
g2o::EdgeSim3ProjectXYZ* e12 = new g2o::EdgeSim3ProjectXYZ();
e12->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id2)));
e12->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(0)));
e12->setMeasurement(obs1);
const float &invSigmaSquare1 = pKF1->mvInvLevelSigma2[kpUn1.octave];
e12->setInformation(Eigen::Matrix2d::Identity()*invSigmaSquare1);
g2o::RobustKernelHuber* rk1 = new g2o::RobustKernelHuber;
e12->setRobustKernel(rk1);
rk1->setDelta(deltaHuber);
optimizer.addEdge(e12);
// Set edge x2 = S21*X1
Eigen::Matrix<double,2,1> obs2;
const cv::KeyPoint &kpUn2 = pKF2->mvKeysUn[i2];
obs2 << kpUn2.pt.x, kpUn2.pt.y;
g2o::EdgeInverseSim3ProjectXYZ* e21 = new g2o::EdgeInverseSim3ProjectXYZ();
e21->setVertex(0, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(id1)));
e21->setVertex(1, dynamic_cast<g2o::OptimizableGraph::Vertex*>(optimizer.vertex(0)));
e21->setMeasurement(obs2);
float invSigmaSquare2 = pKF2->mvInvLevelSigma2[kpUn2.octave];
e21->setInformation(Eigen::Matrix2d::Identity()*invSigmaSquare2);
g2o::RobustKernelHuber* rk2 = new g2o::RobustKernelHuber;
e21->setRobustKernel(rk2);
rk2->setDelta(deltaHuber);
optimizer.addEdge(e21);
vpEdges12.push_back(e12);
vpEdges21.push_back(e21);
vnIndexEdge.push_back(i);
}
// Optimize!
// 步骤3：g2o开始优化，先迭代5次
optimizer.initializeOptimization();
optimizer.optimize(5);
// Check inliers
// 步骤4：剔除一些误差大的边
// 进行卡方检验，大于阈值的边剔除
int nBad=0;
for(size_t i=0; i<vpEdges12.size();i++)
{
g2o::EdgeSim3ProjectXYZ* e12 = vpEdges12[i];
g2o::EdgeInverseSim3ProjectXYZ* e21 = vpEdges21[i];
if(!e12 || !e21)
continue;
if(e12->chi2()>th2 || e21->chi2()>th2)
{
size_t idx = vnIndexEdge[i];
vpMatches1[idx]=static_cast<MapPoint*>(NULL);
optimizer.removeEdge(e12);
optimizer.removeEdge(e21);
vpEdges12[i]=static_cast<g2o::EdgeSim3ProjectXYZ*>(NULL);
vpEdges21[i]=static_cast<g2o::EdgeInverseSim3ProjectXYZ*>(NULL);
nBad++;
}
}
int nMoreIterations;
if(nBad>0)
nMoreIterations=10;
else
nMoreIterations=5;
if(nCorrespondences-nBad<10)
return 0;
// Optimize again only with inliers
// 步骤5：再次g2o优化剔除后剩下的边
optimizer.initializeOptimization();
optimizer.optimize(nMoreIterations);
int nIn = 0;
for(size_t i=0; i<vpEdges12.size();i++)
{
g2o::EdgeSim3ProjectXYZ* e12 = vpEdges12[i];
g2o::EdgeInverseSim3ProjectXYZ* e21 = vpEdges21[i];
if(!e12 || !e21)
continue;
if(e12->chi2()>th2 || e21->chi2()>th2)
{
size_t idx = vnIndexEdge[i];
vpMatches1[idx]=static_cast<MapPoint*>(NULL);
}
else
nIn++;
}
// Recover optimized Sim3
// 步骤6：得到优化后的结果
g2o::VertexSim3Expmap* vSim3_recov = static_cast<g2o::VertexSim3Expmap*>(optimizer.vertex(0));
g2oS12= vSim3_recov->estimate();
return nIn;
}
} //namespace ORB_SLAM