1. SearchByProjection(Frame &F, const vector<MapPoint*> &vpMapPoints, const float th)
| 函数功能 | SearchByProjection函数利用将相机坐标系下的Local MapPoints投影到图像坐标系，在其投影点附近根据描述子距离选取匹配，由此增加当前帧的MapPoints |
| 参数说明 | |
| F | 当前帧 |
| vpMapPoints | Local MapPoints |
| th | 搜索半径 |
| 返回值 | 成功匹配的数量 |
int ORBmatcher::SearchByProjection(Frame &F, const vector<MapPoint*> &vpMapPoints, const float th)
{
int nmatches=0;//匹配数量
const bool bFactor = th!=1.0;//阈值
for(size_t iMP=0; iMP<vpMapPoints.size(); iMP++)//遍历所有MapPoints
{
MapPoint* pMP = vpMapPoints[iMP];
// 判断该点是否要投影
if(!pMP->mbTrackInView)//在SearchLocalPoints()中已经将Local MapPoints
//重投影（isInFrustum()）到当前帧，并标记了这些点是否在当前帧的视野中即mbTrackInView if(pMP->isBad())//如果质量不好不用
continue;
// 通过距离预测的金字塔层数，该层数相对于当前的帧
const int &nPredictedLevel = pMP->mnTrackScaleLevel;
// The size of the window will depend on the viewing direction
// 搜索窗口的大小取决于视角, 若当前视角和平均视角夹角接近0度时, r取一个较小的值
float r = RadiusByViewingCos(pMP->mTrackViewCos);
// 如果需要进行更粗糙的搜索，则增大范围
if(bFactor)
r*=th;
// 通过投影点(投影到当前帧,见isInFrustum())以及搜索窗口和预测的尺度进行搜索, 找出附近的兴趣点
const vector<size_t> vIndices =
F.GetFeaturesInArea(pMP->mTrackProjX,pMP->mTrackProjY,r*F.mvScaleFactors[nPredictedLevel],nPredictedLevel-1,nPredictedLevel);
if(vIndices.empty())//没找到兴趣点
continue;
const cv::Mat MPdescriptor = pMP->GetDescriptor();//求描述子
int bestDist=256;
int bestLevel= -1;
int bestDist2=256;
int bestLevel2 = -1;
int bestIdx =-1 ;
// Get best and second matches with near keypoints
for(vector<size_t>::const_iterator vit=vIndices.begin(), vend=vIndices.end(); vit!=vend; vit++)
{
const size_t idx = *vit;
// 如果Frame中的该兴趣点已经有对应的MapPoint了,则退出该次循环
if(F.mvpMapPoints[idx])
if(F.mvpMapPoints[idx]->Observations()>0)
continue;
if(F.mvuRight[idx]>0)
{
const float er = fabs(pMP->mTrackProjXR-F.mvuRight[idx]);
if(er>r*F.mvScaleFactors[nPredictedLevel])
continue;
}
const cv::Mat &d = F.mDescriptors.row(idx);
const int dist = DescriptorDistance(MPdescriptor,d);//求取描述子距离
// 根据描述子寻找描述子距离最小和次小的特征点
if(dist<bestDist)
{
bestDist2=bestDist;
bestDist=dist;
bestLevel2 = bestLevel;
bestLevel = F.mvKeysUn[idx].octave;
bestIdx=idx;
}
else if(dist<bestDist2)//求次小距离
{
bestLevel2 = F.mvKeysUn[idx].octave;
bestDist2=dist;
}
}
// Apply ratio to second match (only if best and second are in the same scale level)//对第次小距离匹配应用比率
if(bestDist<=TH_HIGH)
{
if(bestLevel==bestLevel2 && bestDist>mfNNratio*bestDist2)
continue;
F.mvpMapPoints[bestIdx]=pMP; // 为Frame中的兴趣点增加对应的MapPoint
nmatches++;
}
}
return nmatches;
}
2. SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)
| 函数功能 | 通过词袋（bow）对关键帧（pKF）和当前帧（F）中的特征点进行快速匹配，不属于同一节点（node）的特征点直接跳过匹配，对属于同一节点（node0的特征点通过描述子距离进行匹配，根据匹配，用关键帧（pKF）中特征点对应的MapPoint更新F中特征点对应的MapPoints，通过距离阈值、比例阈值和角度投票进行剔除误匹配 |
| pKF | 关键帧 |
| F | 当前帧 |
| vpMapPointMatches | 当前帧中MapPoints对应的匹配，NULL表示未匹配 |
| 返回值 | 成功匹配的数量 |
int ORBmatcher::SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)
{
const vector<MapPoint*> vpMapPointsKF = pKF->GetMapPointMatches();
vpMapPointMatches = vector<MapPoint*>(F.N,static_cast<MapPoint*>(NULL));
const DBoW2::FeatureVector &vFeatVecKF = pKF->mFeatVec;
int nmatches=0;//匹配点个数
vector<int> rotHist[HISTO_LENGTH];
for(int i=0;i<HISTO_LENGTH;i++)
rotHist[i].reserve(500);
const float factor = 1.0f/HISTO_LENGTH;
// We perform the matching over ORB that belong to the same vocabulary node (at a certain level)
// 将属于同一节点(特定层)的ORB特征进行匹配
DBoW2::FeatureVector::const_iterator KFit = vFeatVecKF.begin();
DBoW2::FeatureVector::const_iterator Fit = F.mFeatVec.begin();
DBoW2::FeatureVector::const_iterator KFend = vFeatVecKF.end();
DBoW2::FeatureVector::const_iterator Fend = F.mFeatVec.end();
while(KFit != KFend && Fit != Fend)
{
if(KFit->first == Fit->first) //步骤1：分别取出属于同一node的ORB特征点(只有属于同一node，才有可能是匹配点)
{
const vector<unsigned int> vIndicesKF = KFit->second;
const vector<unsigned int> vIndicesF = Fit->second;
// 步骤2：遍历KF中属于该node的特征点
for(size_t iKF=0; iKF<vIndicesKF.size(); iKF++)
{
const unsigned int realIdxKF = vIndicesKF[iKF];
MapPoint* pMP = vpMapPointsKF[realIdxKF]; // 取出KF中该特征对应的MapPoint
if(!pMP)
continue;
if(pMP->isBad())
continue;
const cv::Mat &dKF= pKF->mDescriptors.row(realIdxKF); // 取出KF中该特征对应的描述子
int bestDist1=256; // 最好的距离（最小距离）
int bestIdxF =-1 ;
int bestDist2=256; // 倒数第二好距离（倒数第二小距离）
// 步骤3：遍历F中属于该node的特征点，找到了最佳匹配点
for(size_t iF=0; iF<vIndicesF.size(); iF++)
{
const unsigned int realIdxF = vIndicesF[iF];
if(vpMapPointMatches[realIdxF])// 表明这个点已经被匹配过了，不再匹配，加快速度
continue;
const cv::Mat &dF = F.mDescriptors.row(realIdxF); // 取出F中该特征对应的描述子
const int dist = DescriptorDistance(dKF,dF); // 求描述子的距离
if(dist<bestDist1)// dist < bestDist1 < bestDist2，更新bestDist1 bestDist2
{
bestDist2=bestDist1;
bestDist1=dist;
bestIdxF=realIdxF;
}
else if(dist<bestDist2)// bestDist1 < dist < bestDist2，更新bestDist2
{
bestDist2=dist;
}
}
// 步骤4：根据阈值 和 角度投票剔除误匹配
if(bestDist1<=TH_LOW) // 匹配距离（误差）小于阈值
{
// 最佳匹配比次佳匹配明显要好，那么最佳匹配才真正靠谱
if(static_cast<float>(bestDist1)<mfNNratio*static_cast<float>(bestDist2))
{
// 步骤5：更新特征点的MapPoint
vpMapPointMatches[bestIdxF]=pMP;
const cv::KeyPoint &kp = pKF->mvKeysUn[realIdxKF];
if(mbCheckOrientation)
{
// trick!
// angle：每个特征点在提取描述子时的旋转主方向角度，如果图像旋转了，这个角度将发生改变
// 所有的特征点的角度变化应该是一致的，通过直方图统计得到最准确的角度变化值
float rot = kp.angle-F.mvKeys[bestIdxF].angle;// 该特征点的角度变化值
if(rot<0.0)
rot+=360.0f;
int bin = round(rot*factor);// 将rot分配到bin组
if(bin==HISTO_LENGTH)
bin=0;
assert(bin>=0 && bin<HISTO_LENGTH);//断言函数
rotHist[bin].push_back(bestIdxF);
}
nmatches++;
}
}
}
KFit++;
Fit++;
}
else if(KFit->first < Fit->first)
{
KFit = vFeatVecKF.lower_bound(Fit->first);
}
else
{
Fit = F.mFeatVec.lower_bound(KFit->first);
}
}
// 根据方向剔除误匹配的点
if(mbCheckOrientation)
{
int ind1=-1;
int ind2=-1;
int ind3=-1;
// 计算rotHist中最大的三个的index
ComputeThreeMaxima(rotHist,HISTO_LENGTH,ind1,ind2,ind3);
for(int i=0; i<HISTO_LENGTH; i++)
{
// 如果特征点的旋转角度变化量属于这三个组，则保留
if(i==ind1 || i==ind2 || i==ind3)
continue;
// 将除了ind1 ind2 ind3以外的匹配点去掉
for(size_t j=0, jend=rotHist[i].size(); j<jend; j++)
{
vpMapPointMatches[rotHist[i][j]]=static_cast<MapPoint*>(NULL);
nmatches--;
}
}
}
return nmatches;
}
3. SearchByProjection(KeyFrame* pKF, cv::Mat Scw, const vector<MapPoint*> &vpPoints, vector<MapPoint*> &vpMatched, int th)
| 函数功能 | 根据Sim3变换，将每个vpPoints投影到pKF上，并根据尺度确定一个搜索区域，根据该MapPoint的描述子与该区域内的特征点进行匹配，如果匹配误差小于TH_LOW即匹配成功，更新vpMatched |
| 参数说明 | |
| pKF | 关键帧 |
| Scw | Sim3矩阵，即变换矩阵 |
| vpPoints | MapPoint |
| vpMatched | MapPoint的匹配点 |
| th | 搜索半径的因子 |
| 返回值 | 成功匹配的数量 |
int ORBmatcher::SearchByProjection(KeyFrame* pKF, cv::Mat Scw, const vector<MapPoint*> &vpPoints, vector<MapPoint*> &vpMatched, int th)
{
// Get Calibration Parameters for later projection
//获取相机内参
const float &fx = pKF->fx;
const float &fy = pKF->fy;
const float &cx = pKF->cx;
const float &cy = pKF->cy;
// Decompose Scw
// 分解Scw矩阵，可以自己推一下公式
cv::Mat sRcw = Scw.rowRange(0,3).colRange(0,3);
const float scw = sqrt(sRcw.row(0).dot(sRcw.row(0)));// 计算得到尺度s
cv::Mat Rcw = sRcw/scw;
cv::Mat tcw = Scw.rowRange(0,3).col(3)/scw;// pKF坐标系下，世界坐标系到pKF的位移，方向由世界坐标系指向pKF
cv::Mat Ow = -Rcw.t()*tcw;// 世界坐标系下，pKF到世界坐标系的位移（世界坐标系原点相对pKF的位置），方向由pKF指向世界坐标系
// Set of MapPoints already found in the KeyFrame
// 使用set类型，并去除没有匹配的点，用于快速检索某个MapPoint是否有匹配
set<MapPoint*> spAlreadyFound(vpMatched.begin(), vpMatched.end());
spAlreadyFound.erase(static_cast<MapPoint*>(NULL));//Null代表没匹配
int nmatches=0;//成功匹配个数
// For each Candidate MapPoint Project and Match
// 遍历所有的MapPoints
for(int iMP=0, iendMP=vpPoints.size(); iMP<iendMP; iMP++)
{
MapPoint* pMP = vpPoints[iMP];
// Discard Bad MapPoints and already found
// 丢弃坏的MapPoints和已经匹配上的MapPoints
if(pMP->isBad() || spAlreadyFound.count(pMP))
continue;
// Get 3D Coords.
//获取三维坐标
cv::Mat p3Dw = pMP->GetWorldPos();
// Transform into Camera Coords.
//转化到相机坐标系
cv::Mat p3Dc = Rcw*p3Dw+tcw;
// Depth must be positive
//求得的深度值必须为正
if(p3Dc.at<float>(2)<0.0)
continue;
// Project into Image投影到图片上
const float invz = 1/p3Dc.at<float>(2);
const float x = p3Dc.at<float>(0)*invz;
const float y = p3Dc.at<float>(1)*invz;
const float u = fx*x+cx;
const float v = fy*y+cy;
// Point must be inside the image不在图片上的点去掉
if(!pKF->IsInImage(u,v))
continue;
// Depth must be inside the scale invariance region of the point
// 判断距离是否在尺度协方差范围内
const float maxDistance = pMP->GetMaxDistanceInvariance();
const float minDistance = pMP->GetMinDistanceInvariance();
cv::Mat PO = p3Dw-Ow;
const float dist = cv::norm(PO);
if(dist<minDistance || dist>maxDistance)//超出范围
continue;
// Viewing angle must be less than 60 deg视角必须小于60度
cv::Mat Pn = pMP->GetNormal();
if(PO.dot(Pn)<0.5*dist)
continue;
int nPredictedLevel = pMP->PredictScale(dist,pKF);
// Search in a radius
// 根据尺度确定搜索半径
const float radius = th*pKF->mvScaleFactors[nPredictedLevel];
const vector<size_t> vIndices = pKF->GetFeaturesInArea(u,v,radius);
if(vIndices.empty())
continue;
// Match to the most similar keypoint in the radius匹配最相似的点
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = 256;
int bestIdx = -1;
// 遍历搜索区域内所有特征点，与该MapPoint的描述子进行匹配
for(vector<size_t>::const_iterator vit=vIndices.begin(), vend=vIndices.end(); vit!=vend; vit++)
{
const size_t idx = *vit;
if(vpMatched[idx])
continue;
const int &kpLevel= pKF->mvKeysUn[idx].octave;
if(kpLevel<nPredictedLevel-1 || kpLevel>nPredictedLevel)
continue;
const cv::Mat &dKF = pKF->mDescriptors.row(idx);
const int dist = DescriptorDistance(dMP,dKF);
if(dist<bestDist)//求最小距离
{
bestDist = dist;
bestIdx = idx;
}
}
// 该MapPoint与bestIdx对应的特征点匹配成功
if(bestDist<=TH_LOW)
{
vpMatched[bestIdx]=pMP;
nmatches++;
}
}
return nmatches;
}
4. SearchByBoW(KeyFrame *pKF1, KeyFrame *pKF2, vector<MapPoint *> &vpMatches12)
| 函数功能 | 通过词包，对关键帧的特征点进行跟踪，该函数用于闭环检测时两个关键帧间的特征点匹配，通过bow对pKF和F中的特征点进行快速匹配（不属于同一node的特征点直接跳过匹配），对属于同一node的特征点通过描述子距离进行匹配，根据匹配，更新vpMatches12 |
| pKF1 | KeyFrame1 |
| pKF2 | KeyFrame2 |
| vpMatches12 | pKF2中与pKF1匹配的MapPoint，null表示没有匹配 |
| 返回值 | 成功匹配的数量 |
/**
* @brief 通过语法树加速两个关键帧之间的特征匹配。该函数用于闭环检测时两个关键帧间的特征点匹配
*
* 通过bow对pKF和F中的特征点进行快速匹配（不属于同一node的特征点直接跳过匹配） \n
* 对属于同一node的特征点通过描述子距离进行匹配 \n
* 根据匹配，更新vpMatches12 \n
* 通过距离阈值、比例阈值和角度投票进行剔除误匹配
* @param pKF1 KeyFrame1
* @param pKF2 KeyFrame2
* @param vpMatches12 pKF2中与pKF1匹配的MapPoint，null表示没有匹配
* @return 成功匹配的数量
*/
int ORBmatcher::SearchByBoW(KeyFrame *pKF1, KeyFrame *pKF2, vector<MapPoint *> &vpMatches12)
{
// 详细注释可参见：SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)
const vector<cv::KeyPoint> &vKeysUn1 = pKF1->mvKeysUn;
const DBoW2::FeatureVector &vFeatVec1 = pKF1->mFeatVec;
const vector<MapPoint*> vpMapPoints1 = pKF1->GetMapPointMatches();
const cv::Mat &Descriptors1 = pKF1->mDescriptors;
const vector<cv::KeyPoint> &vKeysUn2 = pKF2->mvKeysUn;
const DBoW2::FeatureVector &vFeatVec2 = pKF2->mFeatVec;
const vector<MapPoint*> vpMapPoints2 = pKF2->GetMapPointMatches();
const cv::Mat &Descriptors2 = pKF2->mDescriptors;
vpMatches12 = vector<MapPoint*>(vpMapPoints1.size(),static_cast<MapPoint*>(NULL));
vector<bool> vbMatched2(vpMapPoints2.size(),false);
vector<int> rotHist[HISTO_LENGTH];
for(int i=0;i<HISTO_LENGTH;i++)
rotHist[i].reserve(500);
const float factor = HISTO_LENGTH/360.0f;
int nmatches = 0;
DBoW2::FeatureVector::const_iterator f1it = vFeatVec1.begin();
DBoW2::FeatureVector::const_iterator f2it = vFeatVec2.begin();
DBoW2::FeatureVector::const_iterator f1end = vFeatVec1.end();
DBoW2::FeatureVector::const_iterator f2end = vFeatVec2.end();
while(f1it != f1end && f2it != f2end)
{
if(f1it->first == f2it->first)//步骤1：分别取出属于同一node的ORB特征点(只有属于同一node，才有可能是匹配点)
{
// 步骤2：遍历KF中属于该node的特征点
for(size_t i1=0, iend1=f1it->second.size(); i1<iend1; i1++)
{
const size_t idx1 = f1it->second[i1];
MapPoint* pMP1 = vpMapPoints1[idx1];
if(!pMP1)
continue;
if(pMP1->isBad())
continue;
const cv::Mat &d1 = Descriptors1.row(idx1);
int bestDist1=256;
int bestIdx2 =-1 ;
int bestDist2=256;
// 步骤3：遍历F中属于该node的特征点，找到了最佳匹配点
for(size_t i2=0, iend2=f2it->second.size(); i2<iend2; i2++)
{
const size_t idx2 = f2it->second[i2];
MapPoint* pMP2 = vpMapPoints2[idx2];
if(vbMatched2[idx2] || !pMP2)
continue;
if(pMP2->isBad())
continue;
const cv::Mat &d2 = Descriptors2.row(idx2);
int dist = DescriptorDistance(d1,d2);
if(dist<bestDist1)
{
bestDist2=bestDist1;
bestDist1=dist;
bestIdx2=idx2;
}
else if(dist<bestDist2)
{
bestDist2=dist;
}
}
// 步骤4：根据阈值 和 角度投票剔除误匹配
// 详见SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)函数步骤4
if(bestDist1<TH_LOW)
{
if(static_cast<float>(bestDist1)<mfNNratio*static_cast<float>(bestDist2))
{
vpMatches12[idx1]=vpMapPoints2[bestIdx2];
vbMatched2[bestIdx2]=true;
if(mbCheckOrientation)
{
float rot = vKeysUn1[idx1].angle-vKeysUn2[bestIdx2].angle;
if(rot<0.0)
rot+=360.0f;
int bin = round(rot*factor);
if(bin==HISTO_LENGTH)
bin=0;
assert(bin>=0 && bin<HISTO_LENGTH);
rotHist[bin].push_back(idx1);
}
nmatches++;
}
}
}
f1it++;
f2it++;
}
else if(f1it->first < f2it->first)
{
f1it = vFeatVec1.lower_bound(f2it->first);
}
else
{
f2it = vFeatVec2.lower_bound(f1it->first);
}
}
if(mbCheckOrientation)
{
int ind1=-1;
int ind2=-1;
int ind3=-1;
ComputeThreeMaxima(rotHist,HISTO_LENGTH,ind1,ind2,ind3);
for(int i=0; i<HISTO_LENGTH; i++)
{
if(i==ind1 || i==ind2 || i==ind3)
continue;
for(size_t j=0, jend=rotHist[i].size(); j<jend; j++)
{
vpMatches12[rotHist[i][j]]=static_cast<MapPoint*>(NULL);
nmatches--;
}
}
}
return nmatches;
}
5. SearchForTriangulation(KeyFrame *pKF1, KeyFrame *pKF2, cv::Mat F12,vector<pair<size_t, size_t> > &vMatchedPairs, const bool bOnlyStereo)
| 函数功能 | 利用基本矩阵F12，在两个关键帧之间未匹配的特征点中产生新的3d点
|
| 参数说明 | |
| pKF1 | 关键帧1 |
| pKF2 | 关键帧2 |
| F12 | 基础矩阵 |
| vMatchedPairs | 存储匹配特征点对，特征点用其在关键帧中的索引表示 |
| bOnlyStereo | 在双目和rgbd情况下，要求特征点在右图存在匹配 |
| 返回值 | 成功匹配的数量 |
int ORBmatcher::SearchForTriangulation(KeyFrame *pKF1, KeyFrame *pKF2, cv::Mat F12,
vector<pair<size_t, size_t> > &vMatchedPairs, const bool bOnlyStereo)
{
const DBoW2::FeatureVector &vFeatVec1 = pKF1->mFeatVec;//关键帧1的特征向量
const DBoW2::FeatureVector &vFeatVec2 = pKF2->mFeatVec;//关键帧2的特征向量
// Compute epipole in second image
// 计算KF1的相机中心在KF2图像平面的坐标，即极点坐标
cv::Mat Cw = pKF1->GetCameraCenter(); // twc1
cv::Mat R2w = pKF2->GetRotation(); // Rc2w
cv::Mat t2w = pKF2->GetTranslation(); // tc2w
cv::Mat C2 = R2w*Cw+t2w; // tc2c1 KF1的相机中心在KF2坐标系的表示
const float invz = 1.0f/C2.at<float>(2);
// 步骤0：得到KF1的相机光心在KF2中的坐标（极点坐标），即投影
const float ex =pKF2->fx*C2.at<float>(0)*invz+pKF2->cx;
const float ey =pKF2->fy*C2.at<float>(1)*invz+pKF2->cy;
// Find matches between not tracked keypoints查找未跟踪关键点之间的匹配
// Matching speed-up by ORB Vocabulary使用词袋加速
// Compare only ORB that share the same node只比较同一节点下的ORB
int nmatches=0;
vector<bool> vbMatched2(pKF2->N,false);
vector<int> vMatches12(pKF1->N,-1);
vector<int> rotHist[HISTO_LENGTH];
for(int i=0;i<HISTO_LENGTH;i++)
rotHist[i].reserve(500);
const float factor = 1.0f/HISTO_LENGTH;
// We perform the matching over ORB that belong to the same vocabulary node (at a certain level)
// 将属于同一节点(特定层)的ORB特征进行匹配
// FeatureVector的数据结构类似于：{(node1,feature_vector1) (node2,feature_vector2)...}
// f1it->first对应node编号，f1it->second对应属于该node的所有特特征点编号
DBoW2::FeatureVector::const_iterator f1it = vFeatVec1.begin();
DBoW2::FeatureVector::const_iterator f2it = vFeatVec2.begin();
DBoW2::FeatureVector::const_iterator f1end = vFeatVec1.end();
DBoW2::FeatureVector::const_iterator f2end = vFeatVec2.end();
// 步骤1：遍历pKF1和pKF2中的node节点
while(f1it!=f1end && f2it!=f2end)
{
// 如果f1it和f2it属于同一个node节点
if(f1it->first == f2it->first)
{
// 步骤2：遍历该node节点下(f1it->first)的所有特征点
for(size_t i1=0, iend1=f1it->second.size(); i1<iend1; i1++)
{
// 获取pKF1中属于该node节点的所有特征点索引
const size_t idx1 = f1it->second[i1];
// 步骤2.1：通过特征点索引idx1在pKF1中取出对应的MapPoint
MapPoint* pMP1 = pKF1->GetMapPoint(idx1);
// If there is already a MapPoint skip
// 由于寻找的是未匹配的特征点，所以pMP1应该为NULL
if(pMP1)
continue;
// 如果mvuRight中的值大于0，表示是双目，且该特征点有深度值
const bool bStereo1 = pKF1->mvuRight[idx1]>=0;
if(bOnlyStereo)
if(!bStereo1)
continue;
// 步骤2.2：通过特征点索引idx1在pKF1中取出对应的特征点
const cv::KeyPoint &kp1 = pKF1->mvKeysUn[idx1];
// 步骤2.3：通过特征点索引idx1在pKF1中取出对应的特征点的描述子
const cv::Mat &d1 = pKF1->mDescriptors.row(idx1);
int bestDist = TH_LOW;
int bestIdx2 = -1;
// 步骤3：遍历该node节点下(f2it->first)的所有特征点
for(size_t i2=0, iend2=f2it->second.size(); i2<iend2; i2++)
{
// 获取pKF2中属于该node节点的所有特征点索引
size_t idx2 = f2it->second[i2];
// 步骤3.1：通过特征点索引idx2在pKF2中取出对应的MapPoint
MapPoint* pMP2 = pKF2->GetMapPoint(idx2);
// If we have already matched or there is a MapPoint skip
// 如果pKF2当前特征点索引idx2已经被匹配过或者对应的3d点非空
// 那么这个索引idx2就不能被考虑
if(vbMatched2[idx2] || pMP2)
continue;
const bool bStereo2 = pKF2->mvuRight[idx2]>=0;
if(bOnlyStereo)
if(!bStereo2)
continue;
// 步骤3.2：通过特征点索引idx2在pKF2中取出对应的特征点的描述子
const cv::Mat &d2 = pKF2->mDescriptors.row(idx2);
// 计算idx1与idx2在两个关键帧中对应特征点的描述子距离
const int dist = DescriptorDistance(d1,d2);
if(dist>TH_LOW || dist>bestDist)
continue;
// 步骤3.3：通过特征点索引idx2在pKF2中取出对应的特征点
const cv::KeyPoint &kp2 = pKF2->mvKeysUn[idx2];
if(!bStereo1 && !bStereo2)
{
const float distex = ex-kp2.pt.x;
const float distey = ey-kp2.pt.y;
// 该特征点距离极点太近，表明kp2对应的MapPoint距离pKF1相机太近
if(distex*distex+distey*distey<100*pKF2->mvScaleFactors[kp2.octave])
continue;
}
// 步骤4：计算特征点kp2到kp1极线（kp1对应pKF2的一条极线）的距离是否小于阈值
if(CheckDistEpipolarLine(kp1,kp2,F12,pKF2))
{
bestIdx2 = idx2;
bestDist = dist;
}
}
// 步骤1、2、3、4总结下来就是：将左图像的每个特征点与右图像同一node节点的所有特征点
// 依次检测，判断是否满足对极几何约束，满足约束就是匹配的特征点
// 详见SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)函数步骤4
if(bestIdx2>=0)
{
const cv::KeyPoint &kp2 = pKF2->mvKeysUn[bestIdx2];
vMatches12[idx1]=bestIdx2;
nmatches++;
if(mbCheckOrientation)
{
float rot = kp1.angle-kp2.angle;
if(rot<0.0)
rot+=360.0f;
int bin = round(rot*factor);
if(bin==HISTO_LENGTH)
bin=0;
assert(bin>=0 && bin<HISTO_LENGTH);
rotHist[bin].push_back(idx1);
}
}
}
f1it++;
f2it++;
}
else if(f1it->first < f2it->first)
{
f1it = vFeatVec1.lower_bound(f2it->first);
}
else
{
f2it = vFeatVec2.lower_bound(f1it->first);
}
}
if(mbCheckOrientation)
{
int ind1=-1;
int ind2=-1;
int ind3=-1;
ComputeThreeMaxima(rotHist,HISTO_LENGTH,ind1,ind2,ind3);
for(int i=0; i<HISTO_LENGTH; i++)
{
if(i==ind1 || i==ind2 || i==ind3)
continue;
for(size_t j=0, jend=rotHist[i].size(); j<jend; j++)
{
vMatches12[rotHist[i][j]]=-1;
nmatches--;
}
}
}
vMatchedPairs.clear();
vMatchedPairs.reserve(nmatches);
for(size_t i=0, iend=vMatches12.size(); i<iend; i++)
{
if(vMatches12[i]<0)
continue;
vMatchedPairs.push_back(make_pair(i,vMatches12[i]));
}
return nmatches;
}
6. Fuse(KeyFrame *pKF, const vector<MapPoint *> &vpMapPoints, const float th)
| 函数功能 | 将MapPoints投影到关键帧pKF中，并判断是否有重复的MapPoints,1.如果MapPoint能匹配关键帧的特征点，并且该点有对应的MapPoint，那么将两个MapPoint合并（选择观测数多的）,2.如果MapPoint能匹配关键帧的特征点，并且该点没有对应的MapPoint，那么为该点添加MapPoint |
| 参数说明 | |
| pKF | 相邻关键帧 |
| vpMapPoints | 当前关键帧的MapPoints |
| th | 搜索半径的因子 |
| 返回值 | 重复MapPoints的数量 |
int ORBmatcher::Fuse(KeyFrame *pKF, const vector<MapPoint *> &vpMapPoints, const float th)
{ //获取旋转R，和平移t
cv::Mat Rcw = pKF->GetRotation();
cv::Mat tcw = pKF->GetTranslation();
//获取相机内参
const float &fx = pKF->fx;
const float &fy = pKF->fy;
const float &cx = pKF->cx;
const float &cy = pKF->cy;
const float &bf = pKF->mbf;
//获取相机中心
cv::Mat Ow = pKF->GetCameraCenter();
int nFused=0;//重复MapPoints的数量
const int nMPs = vpMapPoints.size();
// 遍历所有的MapPoints
for(int i=0; i<nMPs; i++)
{
MapPoint* pMP = vpMapPoints[i];
if(!pMP)
continue;
if(pMP->isBad() || pMP->IsInKeyFrame(pKF))
continue;
cv::Mat p3Dw = pMP->GetWorldPos();//获取MP在世界坐标系3D坐标
cv::Mat p3Dc = Rcw*p3Dw + tcw;//求取MP在相机坐标系下的坐标
// Depth must be positive深度值必须为正
if(p3Dc.at<float>(2)<0.0f)
continue;
const float invz = 1/p3Dc.at<float>(2);//1/z
const float x = p3Dc.at<float>(0)*invz;//这几部都是投影公式，见SLAM14讲86页
const float y = p3Dc.at<float>(1)*invz;
const float u = fx*x+cx;
const float v = fy*y+cy;// 步骤1：得到MapPoint在图像上的投影坐标
// Point must be inside the image
if(!pKF->IsInImage(u,v))//如果Point不在图片内
continue;
const float ur = u-bf*invz;//公式z=fb/d，d=ul-ur，这里ul即u
const float maxDistance = pMP->GetMaxDistanceInvariance();
const float minDistance = pMP->GetMinDistanceInvariance();
cv::Mat PO = p3Dw-Ow;
const float dist3D = cv::norm(PO);
// Depth must be inside the scale pyramid of the image深度必须在图像的尺度金字塔内
if(dist3D<minDistance || dist3D>maxDistance )
continue;
// Viewing angle must be less than 60 deg视角必须小于60度
cv::Mat Pn = pMP->GetNormal();
if(PO.dot(Pn)<0.5*dist3D)
continue;
int nPredictedLevel = pMP->PredictScale(dist3D,pKF);
// Search in a radius
const float radius = th*pKF->mvScaleFactors[nPredictedLevel];// 步骤2：根据MapPoint的深度确定尺度，从而确定搜索范围
const vector<size_t> vIndices = pKF->GetFeaturesInArea(u,v,radius);
if(vIndices.empty())
continue;
// Match to the most similar keypoint in the radius与半径中最相似的关键点匹配
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = 256;
int bestIdx = -1;
for(vector<size_t>::const_iterator vit=vIndices.begin(), vend=vIndices.end(); vit!=vend; vit++)// 步骤3：遍历搜索范围内的features
{
const size_t idx = *vit;
const cv::KeyPoint &kp = pKF->mvKeysUn[idx];
const int &kpLevel= kp.octave;
if(kpLevel<nPredictedLevel-1 || kpLevel>nPredictedLevel)
continue;
// 计算MapPoint投影的坐标与这个区域特征点的距离，如果偏差很大，直接跳过特征点匹配
if(pKF->mvuRight[idx]>=0)
{
// Check reprojection error in stereo
const float &kpx = kp.pt.x;
const float &kpy = kp.pt.y;
const float &kpr = pKF->mvuRight[idx];
const float ex = u-kpx;
const float ey = v-kpy;
const float er = ur-kpr;
const float e2 = ex*ex+ey*ey+er*er;
if(e2*pKF->mvInvLevelSigma2[kpLevel]>7.8)
continue;
}
else
{
const float &kpx = kp.pt.x;
const float &kpy = kp.pt.y;
const float ex = u-kpx;
const float ey = v-kpy;
const float e2 = ex*ex+ey*ey;
// 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(e2*pKF->mvInvLevelSigma2[kpLevel]>5.99)
continue;
}
const cv::Mat &dKF = pKF->mDescriptors.row(idx);
const int dist = DescriptorDistance(dMP,dKF);
if(dist<bestDist)// 找MapPoint在该区域最佳匹配的特征点
{
bestDist = dist;
bestIdx = idx;
}
}
// If there is already a MapPoint replace otherwise add new measurement
if(bestDist<=TH_LOW)// 找到了MapPoint在该区域最佳匹配的特征点
{
MapPoint* pMPinKF = pKF->GetMapPoint(bestIdx);
if(pMPinKF)// 如果这个点有对应的MapPoint
{
if(!pMPinKF->isBad())// 如果这个MapPoint不是bad，选择哪一个呢？哪个观测的多选哪一个
{
if(pMPinKF->Observations()>pMP->Observations())
pMP->Replace(pMPinKF);
else
pMPinKF->Replace(pMP);
}
}
else// 如果这个点没有对应的MapPoint，那么为该点添加MapPoint
{
pMP->AddObservation(pKF,bestIdx);
pKF->AddMapPoint(pMP,bestIdx);
}
nFused++
}
}
return nFused;
}
7. Fuse(KeyFrame *pKF, cv::Mat Scw, const vector<MapPoint *> &vpPoints, float th, vector<MapPoint *> &vpReplacePoint)
| 函数功能 | 投影MapPoints到KeyFrame中，并判断是否有重复的MapPoints |
| 参数说明 | |
| pKF | 相邻关键帧 |
| Scw | Scw为世界坐标系到pKF机体坐标系的Sim3变换，用于将世界坐标系下的vpPoints变换到机体坐标系 |
| vpPoints | 当前关键帧的MapPoints |
| th | 搜索半径的因子 |
| vpReplacePoint | 记录下来需要被替换掉的pMPinKF |
| 返回值 | 重复MapPoints的数量 |
// 投影MapPoints（用Sim3: Scw参数）到KeyFrame中，并判断是否有重复的MapPoints
// Scw为世界坐标系到pKF机体坐标系的Sim3变换，用于将世界坐标系下的vpPoints变换到机体坐标系
int ORBmatcher::Fuse(KeyFrame *pKF, cv::Mat Scw, const vector<MapPoint *> &vpPoints, float th, vector<MapPoint *> &vpReplacePoint)
{
// Get Calibration Parameters for later projection
const float &fx = pKF->fx;
const float &fy = pKF->fy;
const float &cx = pKF->cx;
const float &cy = pKF->cy;
// Decompose Scw
// 将Sim3转化为SE3并分解
cv::Mat sRcw = Scw.rowRange(0,3).colRange(0,3);
const float scw = sqrt(sRcw.row(0).dot(sRcw.row(0)));// 计算得到尺度s
cv::Mat Rcw = sRcw/scw;// 除掉s
cv::Mat tcw = Scw.rowRange(0,3).col(3)/scw;// 除掉s
cv::Mat Ow = -Rcw.t()*tcw;
// Set of MapPoints already found in the KeyFrame
const set<MapPoint*> spAlreadyFound = pKF->GetMapPoints();
int nFused=0;
const int nPoints = vpPoints.size();
// For each candidate MapPoint project and match
// 遍历所有的MapPoints
for(int iMP=0; iMP<nPoints; iMP++)
{
MapPoint* pMP = vpPoints[iMP];
// Discard Bad MapPoints and already found
if(pMP->isBad() || spAlreadyFound.count(pMP))
continue;
// Get 3D Coords.
cv::Mat p3Dw = pMP->GetWorldPos();
// Transform into Camera Coords.
cv::Mat p3Dc = Rcw*p3Dw+tcw;
// Depth must be positive
if(p3Dc.at<float>(2)<0.0f)
continue;
// Project into Image
const float invz = 1.0/p3Dc.at<float>(2);
const float x = p3Dc.at<float>(0)*invz;
const float y = p3Dc.at<float>(1)*invz;
const float u = fx*x+cx;
const float v = fy*y+cy;// 得到MapPoint在图像上的投影坐标
// Point must be inside the image
if(!pKF->IsInImage(u,v))
continue;
// Depth must be inside the scale pyramid of the image
// 根据距离是否在图像合理金字塔尺度范围内和观测角度是否小于60度判断该MapPoint是否正常
const float maxDistance = pMP->GetMaxDistanceInvariance();
const float minDistance = pMP->GetMinDistanceInvariance();
cv::Mat PO = p3Dw-Ow;
const float dist3D = cv::norm(PO);
if(dist3D<minDistance || dist3D>maxDistance)
continue;
// Viewing angle must be less than 60 deg
cv::Mat Pn = pMP->GetNormal();
if(PO.dot(Pn)<0.5*dist3D)
continue;
// Compute predicted scale level
const int nPredictedLevel = pMP->PredictScale(dist3D,pKF);
// Search in a radius
// 计算搜索范围
const float radius = th*pKF->mvScaleFactors[nPredictedLevel];
// 收集pKF在该区域内的特征点
const vector<size_t> vIndices = pKF->GetFeaturesInArea(u,v,radius);
if(vIndices.empty())
continue;
// Match to the most similar keypoint in the radius
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = INT_MAX;
int bestIdx = -1;
for(vector<size_t>::const_iterator vit=vIndices.begin(); vit!=vIndices.end(); vit++)
{
const size_t idx = *vit;
const int &kpLevel = pKF->mvKeysUn[idx].octave;
if(kpLevel<nPredictedLevel-1 || kpLevel>nPredictedLevel)
continue;
const cv::Mat &dKF = pKF->mDescriptors.row(idx);
int dist = DescriptorDistance(dMP,dKF);
if(dist<bestDist)
{
bestDist = dist;
bestIdx = idx;
}
}
// If there is already a MapPoint replace otherwise add new measurement
if(bestDist<=TH_LOW)
{
MapPoint* pMPinKF = pKF->GetMapPoint(bestIdx);
// 如果这个MapPoint已经存在，则替换，
// 这里不能轻易替换（因为MapPoint一般会被很多帧都可以观测到），这里先记录下来，之后调用Replace函数来替换
if(pMPinKF)
{
if(!pMPinKF->isBad())
vpReplacePoint[iMP] = pMPinKF;// 记录下来该pMPinKF需要被替换掉
}
else// 如果这个MapPoint不存在，直接添加
{
pMP->AddObservation(pKF,bestIdx);
pKF->AddMapPoint(pMP,bestIdx);
}
nFused++;
}
}
return nFused;
}
8. SearchBySim3(KeyFrame *pKF1, KeyFrame *pKF2, vector<MapPoint*> &vpMatches12,const float &s12, const cv::Mat &R12, const cv::Mat &t12, const float th)
| 函数功能 | 通过Sim3变换，确定pKF1的特征点在pKF2中的大致区域，同理，确定pKF2的特征点在pKF1中的大致区域，在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12（之前使用SearchByBoW进行特征点匹配时会有漏匹配） |
| 参数说明 | |
| pKF1 | 关键帧1 |
| pKF2 | 关键帧2 |
| vpMatches12 | 存储匹配特征点对，特征点用其在关键帧中的索引表示 |
| s12 | 尺度 |
| R12 | 1到2的旋转矩阵 |
| t12 | 1到2的平移 |
| th | 搜索半径的因子 |
| 返回值 | 重复MapPoints的数量 |
| 步骤 | 1.通过Sim变换，确定pKF1的特征点在pKF2中的大致区域，在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12（之前使用SearchByBoW进行特征点匹配时会有漏匹配） 2.通过Sim变换，确定pKF2的特征点在pKF1中的大致区域，在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12 |
// 通过Sim3变换，确定pKF1的特征点在pKF2中的大致区域，同理，确定pKF2的特征点在pKF1中的大致区域
// 在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12（之前使用SearchByBoW进行特征点匹配时会有漏匹配）
int ORBmatcher::SearchBySim3(KeyFrame *pKF1, KeyFrame *pKF2, vector<MapPoint*> &vpMatches12,
const float &s12, const cv::Mat &R12, const cv::Mat &t12, const float th)
{
// 步骤1：变量初始化
const float &fx = pKF1->fx;
const float &fy = pKF1->fy;
const float &cx = pKF1->cx;
const float &cy = pKF1->cy;
// Camera 1 from world
// 从world到camera的变换
cv::Mat R1w = pKF1->GetRotation();
cv::Mat t1w = pKF1->GetTranslation();
//Camera 2 from world
cv::Mat R2w = pKF2->GetRotation();
cv::Mat t2w = pKF2->GetTranslation();
//Transformation between cameras
cv::Mat sR12 = s12*R12;
cv::Mat sR21 = (1.0/s12)*R12.t();
cv::Mat t21 = -sR21*t12;
const vector<MapPoint*> vpMapPoints1 = pKF1->GetMapPointMatches();
const int N1 = vpMapPoints1.size();
const vector<MapPoint*> vpMapPoints2 = pKF2->GetMapPointMatches();
const int N2 = vpMapPoints2.size();
vector<bool> vbAlreadyMatched1(N1,false);// 用于记录该特征点是否被处理过
vector<bool> vbAlreadyMatched2(N2,false);// 用于记录该特征点是否在pKF1中有匹配
// 步骤2：用vpMatches12更新vbAlreadyMatched1和vbAlreadyMatched2
for(int i=0; i<N1; i++)
{
MapPoint* pMP = vpMatches12[i];
if(pMP)
{
vbAlreadyMatched1[i]=true;// 该特征点已经判断过
int idx2 = pMP->GetIndexInKeyFrame(pKF2);
if(idx2>=0 && idx2<N2)
vbAlreadyMatched2[idx2]=true;// 该特征点在pKF1中有匹配
}
}
vector<int> vnMatch1(N1,-1);
vector<int> vnMatch2(N2,-1);
// Transform from KF1 to KF2 and search
// 步骤3.1：通过Sim变换，确定pKF1的特征点在pKF2中的大致区域，
// 在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12
// （之前使用SearchByBoW进行特征点匹配时会有漏匹配）
for(int i1=0; i1<N1; i1++)
{
MapPoint* pMP = vpMapPoints1[i1];
if(!pMP || vbAlreadyMatched1[i1])// 该特征点已经有匹配点了，直接跳过
continue;
if(pMP->isBad())
continue;
cv::Mat p3Dw = pMP->GetWorldPos();
cv::Mat p3Dc1 = R1w*p3Dw + t1w;// 把pKF1系下的MapPoint从world坐标系变换到camera1坐标系
cv::Mat p3Dc2 = sR21*p3Dc1 + t21;// 再通过Sim3将该MapPoint从camera1变换到camera2坐标系
// Depth must be positive
if(p3Dc2.at<float>(2)<0.0)
continue;
// 投影到camera2图像平面
const float invz = 1.0/p3Dc2.at<float>(2);
const float x = p3Dc2.at<float>(0)*invz;
const float y = p3Dc2.at<float>(1)*invz;
const float u = fx*x+cx;
const float v = fy*y+cy;
// Point must be inside the image
if(!pKF2->IsInImage(u,v))
continue;
const float maxDistance = pMP->GetMaxDistanceInvariance();
const float minDistance = pMP->GetMinDistanceInvariance();
const float dist3D = cv::norm(p3Dc2);
// Depth must be inside the scale invariance region
if(dist3D<minDistance || dist3D>maxDistance )
continue;
// Compute predicted octave
// 预测该MapPoint对应的特征点在图像金字塔哪一层
const int nPredictedLevel = pMP->PredictScale(dist3D,pKF2);
// Search in a radius
// 计算特征点搜索半径
const float radius = th*pKF2->mvScaleFactors[nPredictedLevel];
// 取出该区域内的所有特征点
const vector<size_t> vIndices = pKF2->GetFeaturesInArea(u,v,radius);
if(vIndices.empty())
continue;
// Match to the most similar keypoint in the radius
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = INT_MAX;
int bestIdx = -1;
// 遍历搜索区域内的所有特征点，与pMP进行描述子匹配
for(vector<size_t>::const_iterator vit=vIndices.begin(), vend=vIndices.end(); vit!=vend; vit++)
{
const size_t idx = *vit;
const cv::KeyPoint &kp = pKF2->mvKeysUn[idx];
if(kp.octave<nPredictedLevel-1 || kp.octave>nPredictedLevel)
continue;
const cv::Mat &dKF = pKF2->mDescriptors.row(idx);
const int dist = DescriptorDistance(dMP,dKF);
if(dist<bestDist)
{
bestDist = dist;
bestIdx = idx;
}
}
if(bestDist<=TH_HIGH)
{
vnMatch1[i1]=bestIdx;
}
}
// Transform from KF2 to KF1 and search
// 步骤3.2：通过Sim变换，确定pKF2的特征点在pKF1中的大致区域，
// 在该区域内通过描述子进行匹配捕获pKF1和pKF2之前漏匹配的特征点，更新vpMatches12
// （之前使用SearchByBoW进行特征点匹配时会有漏匹配）
for(int i2=0; i2<N2; i2++)
{
MapPoint* pMP = vpMapPoints2[i2];
if(!pMP || vbAlreadyMatched2[i2])
continue;
if(pMP->isBad())
continue;
cv::Mat p3Dw = pMP->GetWorldPos();
cv::Mat p3Dc2 = R2w*p3Dw + t2w;
cv::Mat p3Dc1 = sR12*p3Dc2 + t12;
// Depth must be positive
if(p3Dc1.at<float>(2)<0.0)
continue;
const float invz = 1.0/p3Dc1.at<float>(2);
const float x = p3Dc1.at<float>(0)*invz;
const float y = p3Dc1.at<float>(1)*invz;
const float u = fx*x+cx;
const float v = fy*y+cy;
// Point must be inside the image
if(!pKF1->IsInImage(u,v))
continue;
const float maxDistance = pMP->GetMaxDistanceInvariance();
const float minDistance = pMP->GetMinDistanceInvariance();
const float dist3D = cv::norm(p3Dc1);
// Depth must be inside the scale pyramid of the image
if(dist3D<minDistance || dist3D>maxDistance)
continue;
// Compute predicted octave
const int nPredictedLevel = pMP->PredictScale(dist3D,pKF1);
// Search in a radius of 2.5*sigma(ScaleLevel)
const float radius = th*pKF1->mvScaleFactors[nPredictedLevel];
const vector<size_t> vIndices = pKF1->GetFeaturesInArea(u,v,radius);
if(vIndices.empty())
continue;
// Match to the most similar keypoint in the radius
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = INT_MAX;
int bestIdx = -1;
for(vector<size_t>::const_iterator vit=vIndices.begin(), vend=vIndices.end(); vit!=vend; vit++)
{
const size_t idx = *vit;
const cv::KeyPoint &kp = pKF1->mvKeysUn[idx];
if(kp.octave<nPredictedLevel-1 || kp.octave>nPredictedLevel)
continue;
const cv::Mat &dKF = pKF1->mDescriptors.row(idx);
const int dist = DescriptorDistance(dMP,dKF);
if(dist<bestDist)
{
bestDist = dist;
bestIdx = idx;
}
}
if(bestDist<=TH_HIGH)
{
vnMatch2[i2]=bestIdx;
}
}
// Check agreement
int nFound = 0;
for(int i1=0; i1<N1; i1++)
{
int idx2 = vnMatch1[i1];
if(idx2>=0)
{
int idx1 = vnMatch2[idx2];
if(idx1==i1)
{
vpMatches12[i1] = vpMapPoints2[idx2];
nFound++;
}
}
}
return nFound;
}
8. SearchByProjection(Frame &CurrentFrame, const Frame &LastFrame, const float th, const bool bMono)
| 函数功能 | 上一帧中包含了MapPoints，对这些MapPoints进行tracking，由此增加当前帧的MapPoints， |
| LastFrame | 上一帧 |
| th | 阈值 |
| bMono | 是否为单目 |
| 返回值 | 成功匹配的数量 |
| 步骤 | 1. 将上一帧的MapPoints投影到当前帧(根据速度模型可以估计当前帧的Tcw) 2. 在投影点附近根据描述子距离选取匹配，以及最终的方向投票机制进行剔除 |
/**
* @brief 对上一帧每个3D点通过投影在小范围内找到和最匹配的2D点。从而实现当前帧CurrentFrame对上一帧LastFrame 3D点的匹配跟踪。用于tracking中前后帧跟踪
*
* 上一帧中包含了MapPoints，对这些MapPoints进行tracking，由此增加当前帧的MapPoints \n
* 1. 将上一帧的MapPoints投影到当前帧(根据速度模型可以估计当前帧的Tcw)
* 2. 在投影点附近根据描述子距离选取匹配，以及最终的方向投票机制进行剔除
* @param CurrentFrame 当前帧
* @param LastFrame 上一帧
* @param th 阈值
* @param bMono 是否为单目
* @return 成功匹配的数量
* @see SearchByBoW()
*/
int ORBmatcher::SearchByProjection(Frame &CurrentFrame, const Frame &LastFrame, const float th, const bool bMono)
{
int nmatches = 0;
// Rotation Histogram (to check rotation consistency)
vector<int> rotHist[HISTO_LENGTH];
for(int i=0;i<HISTO_LENGTH;i++)
rotHist[i].reserve(500);
const float factor = HISTO_LENGTH/360.0f;
const cv::Mat Rcw = CurrentFrame.mTcw.rowRange(0,3).colRange(0,3);
const cv::Mat tcw = CurrentFrame.mTcw.rowRange(0,3).col(3);
const cv::Mat twc = -Rcw.t()*tcw; // twc(w)
const cv::Mat Rlw = LastFrame.mTcw.rowRange(0,3).colRange(0,3);
const cv::Mat tlw = LastFrame.mTcw.rowRange(0,3).col(3); // tlw(l)
// vector from LastFrame to CurrentFrame expressed in LastFrame
const cv::Mat tlc = Rlw*twc+tlw; // Rlw*twc(w) = twc(l), twc(l) + tlw(l) = tlc(l)
// 判断前进还是后退，并以此预测特征点在当前帧所在的金字塔层数
const bool bForward = tlc.at<float>(2)>CurrentFrame.mb && !bMono; // 非单目情况，如果Z大于基线，则表示前进
const bool bBackward = -tlc.at<float>(2)>CurrentFrame.mb && !bMono; // 非单目情况，如果Z小于基线，则表示前进
for(int i=0; i<LastFrame.N; i++)
{
MapPoint* pMP = LastFrame.mvpMapPoints[i];
if(pMP)
{
if(!LastFrame.mvbOutlier[i])
{
// 对上一帧有效的MapPoints进行跟踪
// Project
cv::Mat x3Dw = pMP->GetWorldPos();
cv::Mat x3Dc = Rcw*x3Dw+tcw;
const float xc = x3Dc.at<float>(0);
const float yc = x3Dc.at<float>(1);
const float invzc = 1.0/x3Dc.at<float>(2);
if(invzc<0)
continue;
float u = CurrentFrame.fx*xc*invzc+CurrentFrame.cx;
float v = CurrentFrame.fy*yc*invzc+CurrentFrame.cy;
if(u<CurrentFrame.mnMinX || u>CurrentFrame.mnMaxX)
continue;
if(v<CurrentFrame.mnMinY || v>CurrentFrame.mnMaxY)
continue;
int nLastOctave = LastFrame.mvKeys[i].octave;
// Search in a window. Size depends on scale
float radius = th*CurrentFrame.mvScaleFactors[nLastOctave]; // 尺度越大，搜索范围越大
vector<size_t> vIndices2;
// NOTE 尺度越大,图像越小
// 以下可以这么理解，例如一个有一定面积的圆点，在某个尺度n下它是一个特征点
// 当前进时，圆点的面积增大，在某个尺度m下它是一个特征点，由于面积增大，则需要在更高的尺度下才能检测出来
// 因此m>=n，对应前进的情况，nCurOctave>=nLastOctave。后退的情况可以类推
if(bForward) // 前进,则上一帧兴趣点在所在的尺度nLastOctave<=nCurOctave
vIndices2 = CurrentFrame.GetFeaturesInArea(u,v, radius, nLastOctave);
else if(bBackward) // 后退,则上一帧兴趣点在所在的尺度0<=nCurOctave<=nLastOctave
vIndices2 = CurrentFrame.GetFeaturesInArea(u,v, radius, 0, nLastOctave);
else // 在[nLastOctave-1, nLastOctave+1]中搜索
vIndices2 = CurrentFrame.GetFeaturesInArea(u,v, radius, nLastOctave-1, nLastOctave+1);
if(vIndices2.empty())
continue;
const cv::Mat dMP = pMP->GetDescriptor();
int bestDist = 256;
int bestIdx2 = -1;
// 遍历满足条件的特征点
for(vector<size_t>::const_iterator vit=vIndices2.begin(), vend=vIndices2.end(); vit!=vend; vit++)
{
// 如果该特征点已经有对应的MapPoint了,则退出该次循环
const size_t i2 = *vit;
if(CurrentFrame.mvpMapPoints[i2])
if(CurrentFrame.mvpMapPoints[i2]->Observations()>0)
continue;
if(CurrentFrame.mvuRight[i2]>0)
{
// 双目和rgbd的情况，需要保证右图的点也在搜索半径以内
const float ur = u - CurrentFrame.mbf*invzc;
const float er = fabs(ur - CurrentFrame.mvuRight[i2]);
if(er>radius)
continue;
}
const cv::Mat &d = CurrentFrame.mDescriptors.row(i2);
const int dist = DescriptorDistance(dMP,d);
if(dist<bestDist)
{
bestDist=dist;
bestIdx2=i2;
}
}
// 详见SearchByBoW(KeyFrame* pKF,Frame &F, vector<MapPoint*> &vpMapPointMatches)函数步骤4
if(bestDist<=TH_HIGH)
{
CurrentFrame.mvpMapPoints[bestIdx2]=pMP; // 为当前帧添加MapPoint
nmatches++;
if(mbCheckOrientation)
{
float rot = LastFrame.mvKeysUn[i].angle-CurrentFrame.mvKeysUn[bestIdx2].angle;
if(rot<0.0)
rot+=360.0f;
int bin = round(rot*factor);
if(bin==HISTO_LENGTH)
bin=0;
assert(bin>=0 && bin<HISTO_LENGTH);
rotHist[bin].push_back(bestIdx2);
}
}
}
}
}
//Apply rotation consistency
if(mbCheckOrientation)
{
int ind1=-1;
int ind2=-1;
int ind3=-1;
ComputeThreeMaxima(rotHist,HISTO_LENGTH,ind1,ind2,ind3);
for(int i=0; i<HISTO_LENGTH; i++)
{
if(i!=ind1 && i!=ind2 && i!=ind3)
{
for(size_t j=0, jend=rotHist[i].size(); j<jend; j++)
{
CurrentFrame.mvpMapPoints[rotHist[i][j]]=static_cast<MapPoint*>(NULL);
nmatches--;
}
}
}
}
return nmatches;
}
参考文献
ORB-SLAM2从理论到代码实现（五）：ORBmatcher.cc程序详解_波波菠菜的博客-CSDN博客
扫一扫
667
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
