i/rtabmap
1
0
Fork 0
mirror of https://github.com/introlab/rtabmap.git synced 2026-09-02 01:20:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

Texturing: Added min polygon cluster size parameter

Browse Source
This commit is contained in:
matlabbe
2017-03-04 18:20:26 -05:00
parent 4db957f437
commit e003fb8d35
7 changed files with 132 additions and 90 deletions
Download Patch File Download Diff File Expand all files Collapse all files

154
corelib/src/pcl18/surface/impl/texture_mapping.hpp
View File

@@ -1138,51 +1138,46 @@ pcl::TextureMapping<PointInT>::textureMeshwithMultipleCameras2 (
// af first (idx_pcan < current_cam), check if some of the faces attached to previous cameras occlude the current faces
// then (idx_pcam == current_cam), check for self occlusions. At this stage, we skip faces that were already marked as occluded
// project all faces
int occludedFaces = 0;
std::set<int> occludedFaces;
for (std::map<float, int>::iterator jter=sortedVisibleFaces.begin(); jter!=sortedVisibleFaces.end(); ++jter)
//for (unsigned int idx = 0; idx<visibilityIndices.size(); ++idx)
{
int idx_face = jter->second;
//int idx_face = visibilityIndices[idx];
std::map<int, FaceInfo>::iterator iter= visibleFaces[current_cam].find(idx_face);
if(iter != visibleFaces[current_cam].end())
UASSERT(iter != visibleFaces[current_cam].end());
FaceInfo & info = iter->second;
// face is in the camera's FOV
//get its circumsribed circle
double radius;
pcl::PointXY center;
// getTriangleCircumcenterAndSize (info.uv_coord1, info.uv_coord2, info.uv_coord3, center, radius);
getTriangleCircumcscribedCircleCentroid(info.uv_coord1, info.uv_coord2, info.uv_coord3, center, radius); // this function yields faster results than getTriangleCircumcenterAndSize
// get points inside circ.circle
if (kdtree.radiusSearch (center, radius, idxNeighbors, neighborsSquaredDistance) > 0 )
{
FaceInfo & info = iter->second;
// face is in the camera's FOV
//get its circumsribed circle
double radius;
pcl::PointXY center;
// getTriangleCircumcenterAndSize (info.uv_coord1, info.uv_coord2, info.uv_coord3, center, radius);
getTriangleCircumcscribedCircleCentroid(info.uv_coord1, info.uv_coord2, info.uv_coord3, center, radius); // this function yields faster results than getTriangleCircumcenterAndSize
// get points inside circ.circle
if (kdtree.radiusSearch (center, radius, idxNeighbors, neighborsSquaredDistance) > 0 )
// for each neighbor
for (size_t i = 0; i < idxNeighbors.size (); ++i)
{
// for each neighbor
for (size_t i = 0; i < idxNeighbors.size (); ++i)
int neighborFaceIndex = idxNeighbors[i]/3;
//std::map<int, FaceInfo>::iterator jter= visibleFaces[current_cam].find(visibilityIndices[neighborFaceIndex]);
//if(jter != visibleFaces[current_cam].end())
{
int neighborFaceIndex = idxNeighbors[i]/3;
//std::map<int, FaceInfo>::iterator jter= visibleFaces[current_cam].find(visibilityIndices[neighborFaceIndex]);
//if(jter != visibleFaces[current_cam].end())
if (std::max(camera_cloud->points[faces[idx_face].vertices[0]].z,
std::max (camera_cloud->points[faces[idx_face].vertices[1]].z,
camera_cloud->points[faces[idx_face].vertices[2]].z))
< camera_cloud->points[faces[visibilityIndices[neighborFaceIndex]].vertices[idxNeighbors[i]%3]].z)
//if (info.distance < jter->second.distance)
{
if (std::max(camera_cloud->points[faces[idx_face].vertices[0]].z,
std::max (camera_cloud->points[faces[idx_face].vertices[1]].z,
camera_cloud->points[faces[idx_face].vertices[2]].z))
< camera_cloud->points[faces[visibilityIndices[neighborFaceIndex]].vertices[idxNeighbors[i]%3]].z)
//if (info.distance < jter->second.distance)
// neighbor is farther than all the face's points. Check if it falls into the triangle
if (checkPointInsideTriangle(info.uv_coord1, info.uv_coord2, info.uv_coord3, projections->at(idxNeighbors[i])))
{
// neighbor is farther than all the face's points. Check if it falls into the triangle
if (checkPointInsideTriangle(info.uv_coord1, info.uv_coord2, info.uv_coord3, projections->at(idxNeighbors[i])))
{
// current neighbor is inside triangle and is closer => the corresponding face
if(visibleFaces[current_cam].erase(visibilityIndices[neighborFaceIndex]))
{
++occludedFaces;
}
//TODO we could remove the projections of this face from the kd-tree cloud, but I fond it slower, and I need the point to keep ordered to querry UV coordinates later
}
// current neighbor is inside triangle and is closer => the corresponding face
occludedFaces.insert(visibilityIndices[neighborFaceIndex]);
//TODO we could remove the projections of this face from the kd-tree cloud, but I fond it slower, and I need the point to keep ordered to querry UV coordinates later
}
}
}
@@ -1190,54 +1185,63 @@ pcl::TextureMapping<PointInT>::textureMeshwithMultipleCameras2 (
}
}
// remove occluded faces
for(std::set<int>::iterator iter= occludedFaces.begin(); iter!=occludedFaces.end(); ++iter)
{
visibleFaces[current_cam].erase(*iter);
}
// filter clusters
int clusterFaces = 0;
std::vector<pcl::Vertices> polygons(visibleFaces[current_cam].size());
std::vector<int> polygon_to_face_index(visibleFaces[current_cam].size());
oi =0;
for(std::map<int, FaceInfo>::iterator iter=visibleFaces[current_cam].begin(); iter!=visibleFaces[current_cam].end(); ++iter)
if(min_cluster_size_>0)
{
polygons[oi].vertices.resize(3);
polygons[oi].vertices[0] = faces[iter->first].vertices[0];
polygons[oi].vertices[1] = faces[iter->first].vertices[1];
polygons[oi].vertices[2] = faces[iter->first].vertices[2];
polygon_to_face_index[oi] = iter->first;
++oi;
}
std::vector<std::set<int> > neighbors;
std::vector<std::set<int> > vertexToPolygons;
rtabmap::util3d::createPolygonIndexes(polygons,
(int)camera_cloud->size(),
neighbors,
vertexToPolygons);
std::list<std::list<int> > clusters = rtabmap::util3d::clusterPolygons(
neighbors,
10);
std::set<int> polygonsKept;
for(std::list<std::list<int> >::iterator iter=clusters.begin(); iter!=clusters.end(); ++iter)
{
for(std::list<int>::iterator jter=iter->begin(); jter!=iter->end(); ++jter)
std::vector<pcl::Vertices> polygons(visibleFaces[current_cam].size());
std::vector<int> polygon_to_face_index(visibleFaces[current_cam].size());
oi =0;
for(std::map<int, FaceInfo>::iterator iter=visibleFaces[current_cam].begin(); iter!=visibleFaces[current_cam].end(); ++iter)
{
polygonsKept.insert(polygon_to_face_index[*jter]);
faceCameras[polygon_to_face_index[*jter]].push_back(current_cam);
polygons[oi].vertices.resize(3);
polygons[oi].vertices[0] = faces[iter->first].vertices[0];
polygons[oi].vertices[1] = faces[iter->first].vertices[1];
polygons[oi].vertices[2] = faces[iter->first].vertices[2];
polygon_to_face_index[oi] = iter->first;
++oi;
}
std::vector<std::set<int> > neighbors;
std::vector<std::set<int> > vertexToPolygons;
rtabmap::util3d::createPolygonIndexes(polygons,
(int)camera_cloud->size(),
neighbors,
vertexToPolygons);
std::list<std::list<int> > clusters = rtabmap::util3d::clusterPolygons(
neighbors,
min_cluster_size_);
std::set<int> polygonsKept;
for(std::list<std::list<int> >::iterator iter=clusters.begin(); iter!=clusters.end(); ++iter)
{
for(std::list<int>::iterator jter=iter->begin(); jter!=iter->end(); ++jter)
{
polygonsKept.insert(polygon_to_face_index[*jter]);
faceCameras[polygon_to_face_index[*jter]].push_back(current_cam);
}
}
for(std::map<int, FaceInfo>::iterator iter=visibleFaces[current_cam].begin(); iter!=visibleFaces[current_cam].end();)
{
if(polygonsKept.find(iter->first) == polygonsKept.end())
{
visibleFaces[current_cam].erase(iter++);
++clusterFaces;
}
else
{
++iter;
}
}
}
for(std::map<int, FaceInfo>::iterator iter=visibleFaces[current_cam].begin(); iter!=visibleFaces[current_cam].end();)
{
if(polygonsKept.find(iter->first) == polygonsKept.end())
{
visibleFaces[current_cam].erase(iter++);
++clusterFaces;
}
else
{
++iter;
}
}
std::string msg = uFormat("Processed camera %d/%d: %d occluded and %d spurious polygons out of %d", (int)current_cam+1, (int)cameras.size(), occludedFaces, clusterFaces, (int)visibilityIndices.size());
std::string msg = uFormat("Processed camera %d/%d: %d occluded and %d spurious polygons out of %d", (int)current_cam+1, (int)cameras.size(), (int)occludedFaces.size(), clusterFaces, (int)visibilityIndices.size());
UINFO(msg.c_str());
if(state && !state->callback(msg))
{

11
corelib/src/pcl18/surface/texture_mapping.h
View File

@@ -119,7 +119,7 @@ namespace pcl
/** \brief Constructor. */
TextureMapping () :
f_ (), vector_field_ (), tex_files_ (), tex_material_ (), max_distance_(0.0f)
f_ (), vector_field_ (), tex_files_ (), tex_material_ (), max_distance_(0.0f), min_cluster_size_(50)
{
}
@@ -174,6 +174,12 @@ namespace pcl
max_distance_ = maxDistance;
}
inline void
setMinClusterSize(int size)
{
min_cluster_size_ = size;
}
/** \brief Map texture to a mesh synthesis algorithm
* \param[in] tex_mesh texture mesh
*/
@@ -363,6 +369,9 @@ namespace pcl
/** \brief maximum distance between camera and polygon to apply a texture */
float max_distance_;
/** \brief Remove texture from small polygon clusters */
int min_cluster_size_;
/** \brief Map texture to a face
* \param[in] p1 the first point
* \param[in] p2 the second point

2
corelib/src/util3d_surface.cpp
View File

@@ -608,6 +608,7 @@ pcl::TextureMesh::Ptr createTextureMesh(
const std::map<int, Transform> & poses,
const std::map<int, CameraModel> & cameraModels,
float maxDistance,
int minClusterSize,
const ProgressState * state)
{
UASSERT(mesh->polygons.size());
@@ -665,6 +666,7 @@ pcl::TextureMesh::Ptr createTextureMesh(
// Texture by projection
pcl::TextureMapping<pcl::PointXYZ> tm; // TextureMapping object that will perform the sort
tm.setMaxDistance(maxDistance);
tm.setMinClusterSize(minClusterSize);
if(tm.textureMeshwithMultipleCameras2(*textureMesh, cameras, state))
{
// compute normals for the mesh if not already here