Files
rtabmap/corelib/src/util3d.cpp
T
matlabbe ee49beaf4f Adding doc and tests (#1492)
* added doc and tests for util2d.h

* updated cmake-ros ci

* Added util3d.h doc and tests

* util3d_transforms.h: Added doc and tests

* util3d_filtering.h: started doc and test

* util3d_filtering.h: more tests and doc

* Added more doc/tests

* finished util3d_filtering doc and tests

* added test for util2d::depthBleedingFiltering

* Added util3d_registration tests

* Added util3d_features.h doc/tests

* added doc/tests for util3d_correspondences.h

* added doc/gtest for util3d_mapping.h (missing hpp functions)

* finished testing util3d_mapping.hpp

* Added util3d_motion_estimation.h tests (2D->3D done)

* finished util3d_motion_estimation.h tests

* minimal util3d_surface.h

* Added Transform and VisualWord tests

* Added doc for CameraModel and StereoCameraModel

* Added more logs in ros ci

* Passing tests on fical

* improved all devcontainer

* added devcontainer kilted, fixed source setup.bash, removed ldconfig in ros-cmake workflow

* cleanup

* source ros

* Added utilite tests

* Added testing to appveyor, github actions cancellable on re-commit on same branch

* appveyor testing without all targets

* appveyor: specifying ALL_BUILD target

* Fixed Util2dTest.NMSImageBoundsRespected test

* Fixing PCL Indices error on old pcl

* Added VWDictionary tests and doc. Fixed LSH not working (fix from https://github.com/flann-lib/flann/pull/472

* fixing some appveyor CI errors, added test to check dictionary serialization against all type

* Added StereoDense, StereoBM and StereoSGBM doc and tests

* Added Stereo tests

* Added CameraModel and StereoCameraModel tests

* Added doc and test for Statistics

* Added doc/tests for Signature

* Added doc/test for SensorEvent, added doc for SensorCaptureInfo

* Added doc to SensorData

* Added SensorData tests

* Added SensorCapture and SensorCaptureThread doc and tests

* fixed sensordata test

* updated SSC test and doc

* Added doc and tests for BayesFilter class

* Enabled testing on mac, updated windows testing like on linux

* added test_link

* fixed unresolved on windows

* fixed ThreadHandle error on macos ci

* Added GPS and GeodeticCoords tests

* Added tests for compression

* Added Odometry tests (base class only)

* Added DBDriver tests

* Added coverage report

* uniformized test names

* fixing concurancy and coverage ci

* dont built tools, examples and app for coverage build

* fixed report tool rebuilt without qt compilation error

* updated coverage option

* updated coverage config

* added doc CI job

* fixing windows and mac ci errors

* Added DBDriverSqlite3 tests

* Added IMU tests

* Added Graph tests

* fixing flaky macos test

* Added IMUThread and IMUFilter tests

* Added Landmarks tests

* Added LASWriter tests

* fixing seed flaky test

* fixing flaky macos timing tests

* Added LocalGrid tests

* Added LocalGridMaker tests

* fixing ci errors

* Added GlobalMap tests

* Added doc for EnvSensor

* Added Features2D tests

* Added Registration tests

* Added RegistrationVis tests

* Added doc for Rtabmap and Memory classes

* Added Memory and Rtabmap tests

* making some tests less flaky

* lcov 1.14 support

* updated compatible tool arguments

* Added integration tests (RGB-D, Stereo, Lidar2d, Lidar3d)

* More octomap checks

* Refactored how/when python interpretor is created to simplify library usage

* Added python tests

* fixed some flaky tests

* suppressed some third party related warnings

* fixed ceres tests

* more flaky fixes

* Fixing tests without libpointmatcher

* Added RANSAC rejection filter to PCL ICP

* fixing multi platform flakiness

* Added test to detect regression

* Fixing windows pcl link error

* fixed some macos flakiness

* bigger 2D2D registration error on opencv 4.6.0

* flakiness

* fixing flaky tests on windows and mac

* flaky thread test on slow mac VM

* windows slow test

* fixing more ci erros

* fxing temp dir on windows

* Added Optimizer tests and discovered some bugs (fixed)

* fixing flaky tests in mac and windows

* Added Optimizer doc

* Added GTSAM BA, updated Ceres to use g2o ba parameters. Renamed g2o's ba related parameters to Optimizer group and used by both gtsam and ceres.

* fixing build without gtsam

* fixing home dir

* fixing python ci isssues

* Added multicam ba tests

* Added Ceres multicam BA support

* Aligned BundleAdjustment parameters with Optimizer/Strategy to avoid confusion in the code

* Added BA integration test

* Added robust graph optimization integration test

* Added loop3it test

* Added stereo20Hz test

* Added smartfactor gtsam

* Fixed bugged check and warn if python didn't return any descriptors

* Fixing gtsam version build issues

* fixing tilt on windows ci

* loosing ceres integration test for ci

* mac ci flakiness

* updating missing param in gui

* updating test bound for mac

* added appearance-based tests, set min gftt quality to quality level

* testing more stuff

* improving features2d tests

* ci flakiness

* fixing flaky ci

* ci fixes

* flaky fixes

* Added RegistrationIcp tests

* Added icp integration test with real-worl corridor like env

* intermediate nodes

* fixing enum

* Updated test to catch #1714

* Fixed 2d corridor failing on pcl

* flaky pnp test

* flaky brisk test

* Set rtabmap_integration test as long

* updating loop closure test

* flaky ci tests

* TEsting roundtrip g2o/toro save/load

* loosing test bound

* fixed cuda capable checks

* flaky tests

* Debugging test hanging

* more debugging stuff

* updating limit

* windows: disabled cuda on ci to avoid incompatible driver issue. Fixing a bad test mem allocation

* trying fixing cuda hanging issue

* fixing ci flakyness

* flaky tests

* Updated BOW flaky tests by checking min precision/recall instead of recall@100precision. Fixed signature test

* CameraModel::load() test initRectificationMap param

* test dbdriver load dictionary idsOnly

* Memory: test keepLinkedInDb param

* added dummyDictionary tests

* test intermediate nodes count

* Added MarkerDetector tests

* reverted breaking change of UMutex and USemaphore

* Features2d: fixed compiltion warnings with clang about override

* clang warnings

* fixing test build with pcl 1.8

* g2o and gtsam build errors on android

* opencv5 test fixes

* disabled testing for ios and android builds

* normalized endline characters for easier diff

* added LF CRLF rule

* bump 0.23.10. fixing doc version

* Publish rtabmap website doc from ci

* fixing MSCVC build error

* macos icp flaky test

* fixing ceres macos test bound

* ficing more flaky tests

* fixing opencv5 related test errors. Also fixed an actual bug in ENU_WGS84ToGeocentric_WGS84()

* added comment about mrpt change

* removed rosdoc2 (will add it for rtabmap_ros later)

* fixing website style

* updated download links

* locally deployable website with api

* sweep doxygen issues

* improved/revised doxygen main pages

* removed examples empty page

* Updated doxygen style

* more concise doxygen groups

* added api link on main readme

* fixing utilite test error

* fixing CommonFilteringGroundNormalsUp test

* updated precisionRecall test bounds for Freak and brief descriptors

* fixing scale check in ba tests

* disabled tests on windows cuda build (missing dlls amd runner cannot test cuda anyway)

* ceres: missing suitesparse dep in windows ci

* adjusting recall thr for fast/freak

* ficing more flaky tests

* fixing flaky tests

* disabled coverage in ros ci

* Enable integration tests for ros ci jobs

* loosing up some threshold for failing tests

* trigger cache

* fixing test data in ros ci. Updated flaky test for mac

* slaking some test limit

* Fixed rtabmap-detectMoreLoopClosures inverted output value

* loosing up sift recall on mac

* optimizer re-ordered distribution for reproducible results (mac g2o)

* macos dump test crash log

* combining all tests to save time on shared library reload. Also fixed Logs with missing arguments.

* Added ENABLE_FORMAT_ERRORS cmake option

* do test only one time

* fixed all format warnings

* format security android build errors

* less verbose tests

* updated ImuUThread test

* fixed a log

* Fixed libpointmatcher 2d normals eigen issue

* Fixing libpointmatcher conversion issues

* fixing libpointmatcher test on windows ci

* cleanup comments, relax some test thr

* disabled sequoia-intel ci build (too flaky, would need extensive testing directly on that machine)
2026-08-06 13:32:20 -07:00

3954 lines
112 KiB
C++

/*
Copyright (c) 2010-2016, Mathieu Labbe - IntRoLab - Universite de Sherbrooke
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of the Universite de Sherbrooke nor the
names of its contributors may be used to endorse or promote products
derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <rtabmap/core/util3d.h>
#include <rtabmap/core/util3d_transforms.h>
#include <rtabmap/core/util3d_filtering.h>
#include <rtabmap/core/util3d_surface.h>
#include <rtabmap/core/util2d.h>
#include <rtabmap/core/util2d.h>
#include <rtabmap/utilite/ULogger.h>
#include <rtabmap/utilite/UMath.h>
#include <rtabmap/utilite/UConversion.h>
#include <rtabmap/utilite/UFile.h>
#include <rtabmap/utilite/UTimer.h>
#include <pcl/io/pcd_io.h>
#include <pcl/io/ply_io.h>
#include <pcl/common/transforms.h>
#include <pcl/common/common.h>
#include <opencv2/imgproc/imgproc.hpp>
namespace rtabmap
{
namespace util3d
{
cv::Mat rgbFromCloud(const pcl::PointCloud<pcl::PointXYZRGBA> & cloud, bool bgrOrder)
{
cv::Mat frameBGR = cv::Mat(cloud.height,cloud.width,CV_8UC3);
for(unsigned int h = 0; h < cloud.height; h++)
{
for(unsigned int w = 0; w < cloud.width; w++)
{
if(bgrOrder)
{
frameBGR.at<cv::Vec3b>(h,w)[0] = cloud.at(h*cloud.width + w).b;
frameBGR.at<cv::Vec3b>(h,w)[1] = cloud.at(h*cloud.width + w).g;
frameBGR.at<cv::Vec3b>(h,w)[2] = cloud.at(h*cloud.width + w).r;
}
else
{
frameBGR.at<cv::Vec3b>(h,w)[0] = cloud.at(h*cloud.width + w).r;
frameBGR.at<cv::Vec3b>(h,w)[1] = cloud.at(h*cloud.width + w).g;
frameBGR.at<cv::Vec3b>(h,w)[2] = cloud.at(h*cloud.width + w).b;
}
}
}
return frameBGR;
}
// return float image in meter
cv::Mat depthFromCloud(
const pcl::PointCloud<pcl::PointXYZRGBA> & cloud,
bool depth16U)
{
cv::Mat frameDepth = cv::Mat(cloud.height,cloud.width,depth16U?CV_16UC1:CV_32FC1);
for(unsigned int h = 0; h < cloud.height; h++)
{
for(unsigned int w = 0; w < cloud.width; w++)
{
float depth = cloud.at(h*cloud.width + w).z;
if(depth16U)
{
depth *= 1000.0f;
unsigned short depthMM = 0;
if(depth <= (float)USHRT_MAX)
{
depthMM = (unsigned short)depth;
}
frameDepth.at<unsigned short>(h,w) = depthMM;
}
else
{
frameDepth.at<float>(h,w) = depth;
}
}
}
return frameDepth;
}
// return (unsigned short 16bits image in mm) (float 32bits image in m)
void rgbdFromCloud(const pcl::PointCloud<pcl::PointXYZRGBA> & cloud,
cv::Mat & frameBGR,
cv::Mat & frameDepth,
bool bgrOrder,
bool depth16U)
{
frameDepth = cv::Mat(cloud.height,cloud.width,depth16U?CV_16UC1:CV_32FC1);
frameBGR = cv::Mat(cloud.height,cloud.width,CV_8UC3);
for(unsigned int h = 0; h < cloud.height; h++)
{
for(unsigned int w = 0; w < cloud.width; w++)
{
//rgb
if(bgrOrder)
{
frameBGR.at<cv::Vec3b>(h,w)[0] = cloud.at(h*cloud.width + w).b;
frameBGR.at<cv::Vec3b>(h,w)[1] = cloud.at(h*cloud.width + w).g;
frameBGR.at<cv::Vec3b>(h,w)[2] = cloud.at(h*cloud.width + w).r;
}
else
{
frameBGR.at<cv::Vec3b>(h,w)[0] = cloud.at(h*cloud.width + w).r;
frameBGR.at<cv::Vec3b>(h,w)[1] = cloud.at(h*cloud.width + w).g;
frameBGR.at<cv::Vec3b>(h,w)[2] = cloud.at(h*cloud.width + w).b;
}
//depth
float depth = cloud.at(h*cloud.width + w).z;
if(depth16U)
{
depth *= 1000.0f;
unsigned short depthMM = 0;
if(depth <= (float)USHRT_MAX)
{
depthMM = (unsigned short)depth;
}
frameDepth.at<unsigned short>(h,w) = depthMM;
}
else
{
frameDepth.at<float>(h,w) = depth;
}
}
}
}
pcl::PointXYZ projectDepthTo3D(
const cv::Mat & depthImage,
float x, float y,
float cx, float cy,
float fx, float fy,
bool smoothing,
float depthErrorRatio)
{
UASSERT(depthImage.type() == CV_16UC1 || depthImage.type() == CV_32FC1);
pcl::PointXYZ pt;
float depth = util2d::getDepth(depthImage, x, y, smoothing, depthErrorRatio);
if(depth > 0.0f)
{
// Use correct principal point from calibration
cx = cx > 0.0f ? cx : float(depthImage.cols/2) - 0.5f; //cameraInfo.K.at(2)
cy = cy > 0.0f ? cy : float(depthImage.rows/2) - 0.5f; //cameraInfo.K.at(5)
// Fill in XYZ
pt.x = (x - cx) * depth / fx;
pt.y = (y - cy) * depth / fy;
pt.z = depth;
}
else
{
pt.x = pt.y = pt.z = std::numeric_limits<float>::quiet_NaN();
}
return pt;
}
Eigen::Vector3f projectDepthTo3DRay(
const cv::Size & imageSize,
float x, float y,
float cx, float cy,
float fx, float fy)
{
Eigen::Vector3f ray;
// Use correct principal point from calibration
cx = cx > 0.0f ? cx : float(imageSize.width/2) - 0.5f; //cameraInfo.K.at(2)
cy = cy > 0.0f ? cy : float(imageSize.height/2) - 0.5f; //cameraInfo.K.at(5)
// Fill in XYZ
ray[0] = (x - cx) / fx;
ray[1] = (y - cy) / fy;
ray[2] = 1.0f;
return ray;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloudFromDepth(
const cv::Mat & imageDepth,
float cx, float cy,
float fx, float fy,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
CameraModel model(fx, fy, cx, cy);
return cloudFromDepth(imageDepth, model, decimation, maxDepth, minDepth, validIndices);
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloudFromDepth(
const cv::Mat & imageDepthIn,
const CameraModel & model,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
return cloudFromDepth(
imageDepthIn,
cv::Mat(),
model,
decimation,
maxDepth,
minDepth,
0,
validIndices);
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloudFromDepth(
const cv::Mat & imageDepthIn,
const cv::Mat & imageDepthConfidenceIn,
const CameraModel & model,
int decimation,
float maxDepth,
float minDepth,
unsigned char confidenceThr,
std::vector<int> * validIndices)
{
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
if(decimation == 0)
{
decimation = 1;
}
float rgbToDepthFactorX = 1.0f;
float rgbToDepthFactorY = 1.0f;
UASSERT(model.isValidForProjection());
UASSERT(!imageDepthIn.empty() && (imageDepthIn.type() == CV_16UC1 || imageDepthIn.type() == CV_32FC1));
UASSERT(imageDepthConfidenceIn.empty() || confidenceThr == 0 || (imageDepthConfidenceIn.type() == CV_8UC1 && imageDepthConfidenceIn.size() == imageDepthIn.size()));
cv::Mat imageDepth = imageDepthIn;
cv::Mat imageDepthConfidence = confidenceThr==0?cv::Mat():imageDepthConfidenceIn;
if(model.imageHeight()>0 && model.imageWidth()>0)
{
UASSERT(model.imageHeight() % imageDepthIn.rows == 0 && model.imageWidth() % imageDepthIn.cols == 0);
if(decimation < 0)
{
UDEBUG("Decimation from model (%d)", decimation);
if(model.imageHeight() % decimation != 0)
{
UERROR("Decimation is not valid for current image size (model.imageHeight()=%d decimation=%d). The cloud is not created.", model.imageHeight(), decimation);
return cloud;
}
if(model.imageWidth() % decimation != 0)
{
UERROR("Decimation is not valid for current image size (model.imageWidth()=%d decimation=%d). The cloud is not created.", model.imageWidth(), decimation);
return cloud;
}
// decimate from RGB image size, upsample depth if needed
decimation = -1*decimation;
int targetSize = model.imageHeight() / decimation;
if(targetSize > imageDepthIn.rows)
{
UDEBUG("Depth interpolation factor=%d", targetSize/imageDepthIn.rows);
imageDepth = util2d::interpolate(imageDepthIn, targetSize/imageDepthIn.rows);
if(!imageDepthConfidence.empty()) {
imageDepthConfidence = util2d::interpolate(imageDepthConfidenceIn, targetSize/imageDepthConfidenceIn.rows);
}
decimation = 1;
}
else if(targetSize == imageDepthIn.rows)
{
decimation = 1;
}
else
{
UASSERT(imageDepthIn.rows % targetSize == 0);
decimation = imageDepthIn.rows / targetSize;
}
}
else
{
if(imageDepthIn.rows % decimation != 0)
{
UERROR("Decimation is not valid for current image size (imageDepth.rows=%d decimation=%d). The cloud is not created.", imageDepthIn.rows, decimation);
return cloud;
}
if(imageDepthIn.cols % decimation != 0)
{
UERROR("Decimation is not valid for current image size (imageDepth.cols=%d decimation=%d). The cloud is not created.", imageDepthIn.cols, decimation);
return cloud;
}
}
rgbToDepthFactorX = 1.0f/float((model.imageWidth() / imageDepth.cols));
rgbToDepthFactorY = 1.0f/float((model.imageHeight() / imageDepth.rows));
}
else
{
decimation = abs(decimation);
UASSERT_MSG(imageDepth.rows % decimation == 0, uFormat("rows=%d decimation=%d", imageDepth.rows, decimation).c_str());
UASSERT_MSG(imageDepth.cols % decimation == 0, uFormat("cols=%d decimation=%d", imageDepth.cols, decimation).c_str());
}
//cloud.header = cameraInfo.header;
cloud->height = imageDepth.rows/decimation;
cloud->width = imageDepth.cols/decimation;
cloud->is_dense = false;
cloud->resize(cloud->height * cloud->width);
if(validIndices)
{
validIndices->resize(cloud->size());
}
float depthFx = model.fx() * rgbToDepthFactorX;
float depthFy = model.fy() * rgbToDepthFactorY;
float depthCx = model.cx() * rgbToDepthFactorX;
float depthCy = model.cy() * rgbToDepthFactorY;
UDEBUG("depth=%dx%d fx=%f fy=%f cx=%f cy=%f (depth factors=%f %f) has confidence=%d (thr=%d) decimation=%d",
imageDepth.cols, imageDepth.rows,
model.fx(), model.fy(), model.cx(), model.cy(),
rgbToDepthFactorX,
rgbToDepthFactorY,
imageDepthConfidenceIn.empty()?0:1,
(int)confidenceThr,
decimation);
int oi = 0;
for(int h = 0; h < imageDepth.rows && h/decimation < (int)cloud->height; h+=decimation)
{
for(int w = 0; w < imageDepth.cols && w/decimation < (int)cloud->width; w+=decimation)
{
pcl::PointXYZ & pt = cloud->at((h/decimation)*cloud->width + (w/decimation));
pt.x = pt.y = pt.z = std::numeric_limits<float>::quiet_NaN();
if(imageDepthConfidence.empty() || imageDepthConfidence.at<unsigned char>(h,w) >= confidenceThr)
{
pcl::PointXYZ ptXYZ = projectDepthTo3D(imageDepth, w, h, depthCx, depthCy, depthFx, depthFy, false);
if(pcl::isFinite(ptXYZ) && ptXYZ.z>=minDepth && (maxDepth<=0.0f || ptXYZ.z <= maxDepth))
{
pt.x = ptXYZ.x;
pt.y = ptXYZ.y;
pt.z = ptXYZ.z;
if(validIndices)
{
validIndices->at(oi++) = (h/decimation)*cloud->width + (w/decimation);
}
}
}
}
}
if(validIndices)
{
validIndices->resize(oi);
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudFromDepthRGB(
const cv::Mat & imageRgb,
const cv::Mat & imageDepth,
float cx, float cy,
float fx, float fy,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
CameraModel model(fx, fy, cx, cy);
return cloudFromDepthRGB(imageRgb, imageDepth, model, decimation, maxDepth, minDepth, validIndices);
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudFromDepthRGB(
const cv::Mat & imageRgb,
const cv::Mat & imageDepthIn,
const CameraModel & model,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
return cloudFromDepthRGB(
imageRgb,
imageDepthIn,
cv::Mat(),
model,
decimation,
maxDepth,
minDepth,
0,
validIndices);
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudFromDepthRGB(
const cv::Mat & imageRgb,
const cv::Mat & imageDepthIn,
const cv::Mat & imageDepthConfidenceIn,
const CameraModel & model,
int decimation,
float maxDepth,
float minDepth,
unsigned char confidenceThr,
std::vector<int> * validIndices)
{
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZRGB>);
if(decimation == 0)
{
decimation = 1;
}
UDEBUG("");
UASSERT(model.isValidForProjection());
UASSERT_MSG((model.imageHeight() == 0 && model.imageWidth() == 0) ||
(model.imageHeight() == imageRgb.rows && model.imageWidth() == imageRgb.cols),
uFormat("model=%dx%d rgb=%dx%d", model.imageWidth(), model.imageHeight(), imageRgb.cols, imageRgb.rows).c_str());
//UASSERT_MSG(imageRgb.rows % imageDepthIn.rows == 0 && imageRgb.cols % imageDepthIn.cols == 0,
// uFormat("rgb=%dx%d depth=%dx%d", imageRgb.cols, imageRgb.rows, imageDepthIn.cols, imageDepthIn.rows).c_str());
UASSERT(!imageDepthIn.empty() && (imageDepthIn.type() == CV_16UC1 || imageDepthIn.type() == CV_32FC1));
UASSERT(imageDepthConfidenceIn.empty() || confidenceThr==0 || (imageDepthConfidenceIn.type() == CV_8UC1 && imageDepthConfidenceIn.size() == imageDepthIn.size()));
if(decimation < 0)
{
if(imageRgb.rows % decimation != 0 || imageRgb.cols % decimation != 0)
{
int oldDecimation = decimation;
while(decimation <= -1)
{
if(imageRgb.rows % decimation == 0 && imageRgb.cols % decimation == 0)
{
break;
}
++decimation;
}
if(imageRgb.rows % oldDecimation != 0 || imageRgb.cols % oldDecimation != 0)
{
UWARN("Decimation (%d) is not valid for current image size (rgb=%dx%d). Highest compatible decimation used=%d.", oldDecimation, imageRgb.cols, imageRgb.rows, decimation);
}
}
}
else
{
if(imageDepthIn.rows % decimation != 0 || imageDepthIn.cols % decimation != 0)
{
int oldDecimation = decimation;
while(decimation >= 1)
{
if(imageDepthIn.rows % decimation == 0 && imageDepthIn.cols % decimation == 0)
{
break;
}
--decimation;
}
if(imageDepthIn.rows % oldDecimation != 0 || imageDepthIn.cols % oldDecimation != 0)
{
UWARN("Decimation (%d) is not valid for current image size (depth=%dx%d). Highest compatible decimation used=%d.", oldDecimation, imageDepthIn.cols, imageDepthIn.rows, decimation);
}
}
}
cv::Mat imageDepth = imageDepthIn;
cv::Mat imageDepthConfidence = confidenceThr==0?cv::Mat():imageDepthConfidenceIn;
if(decimation < 0)
{
UDEBUG("Decimation from RGB image (%d)", decimation);
// decimate from RGB image size, upsample depth if needed
decimation = -1*decimation;
int targetSize = imageRgb.rows / decimation;
if(targetSize > imageDepthIn.rows)
{
UDEBUG("Depth interpolation factor=%d", targetSize/imageDepthIn.rows);
imageDepth = util2d::interpolate(imageDepthIn, targetSize/imageDepthIn.rows);
if(!imageDepthConfidence.empty()) {
imageDepthConfidence = util2d::interpolate(imageDepthConfidenceIn, targetSize/imageDepthConfidenceIn.rows);
}
decimation = 1;
}
else if(targetSize == imageDepthIn.rows)
{
decimation = 1;
}
else
{
UASSERT(imageDepthIn.rows % targetSize == 0);
decimation = imageDepthIn.rows / targetSize;
}
}
bool mono;
if(imageRgb.channels() == 3) // BGR
{
mono = false;
}
else if(imageRgb.channels() == 1) // Mono
{
mono = true;
}
else
{
return cloud;
}
//cloud.header = cameraInfo.header;
cloud->height = imageDepth.rows/decimation;
cloud->width = imageDepth.cols/decimation;
cloud->is_dense = false;
cloud->resize(cloud->height * cloud->width);
if(validIndices)
{
validIndices->resize(cloud->size());
}
float rgbToDepthFactorX = float(imageRgb.cols) / float(imageDepth.cols);
float rgbToDepthFactorY = float(imageRgb.rows) / float(imageDepth.rows);
float depthFx = model.fx() / rgbToDepthFactorX;
float depthFy = model.fy() / rgbToDepthFactorY;
float depthCx = model.cx() / rgbToDepthFactorX;
float depthCy = model.cy() / rgbToDepthFactorY;
UDEBUG("rgb=%dx%d depth=%dx%d fx=%f fy=%f cx=%f cy=%f (depth factors=%f %f) has confidence=%d (thr=%d) decimation=%d",
imageRgb.cols, imageRgb.rows,
imageDepth.cols, imageDepth.rows,
model.fx(), model.fy(), model.cx(), model.cy(),
rgbToDepthFactorX,
rgbToDepthFactorY,
imageDepthConfidenceIn.empty()?0:1,
(int)confidenceThr,
decimation);
int oi = 0;
for(int h = 0; h < imageDepth.rows && h/decimation < (int)cloud->height; h+=decimation)
{
for(int w = 0; w < imageDepth.cols && w/decimation < (int)cloud->width; w+=decimation)
{
pcl::PointXYZRGB & pt = cloud->at((h/decimation)*cloud->width + (w/decimation));
int x = int(w*rgbToDepthFactorX);
int y = int(h*rgbToDepthFactorY);
UASSERT(x >=0 && x<imageRgb.cols && y >=0 && y<imageRgb.rows);
if(!mono)
{
const unsigned char * bgr = imageRgb.ptr<unsigned char>(y,x);
pt.b = bgr[0];
pt.g = bgr[1];
pt.r = bgr[2];
}
else
{
unsigned char v = imageRgb.at<unsigned char>(y,x);
pt.b = v;
pt.g = v;
pt.r = v;
}
pt.x = pt.y = pt.z = std::numeric_limits<float>::quiet_NaN();
if(imageDepthConfidence.empty() || imageDepthConfidence.at<unsigned char>(h,w) >= confidenceThr)
{
pcl::PointXYZ ptXYZ = projectDepthTo3D(imageDepth, w, h, depthCx, depthCy, depthFx, depthFy, false);
if (pcl::isFinite(ptXYZ) && ptXYZ.z >= minDepth && (maxDepth <= 0.0f || ptXYZ.z <= maxDepth))
{
pt.x = ptXYZ.x;
pt.y = ptXYZ.y;
pt.z = ptXYZ.z;
if (validIndices)
{
validIndices->at(oi) = (h / decimation)*cloud->width + (w / decimation);
}
++oi;
}
}
}
}
if(validIndices)
{
validIndices->resize(oi);
}
if(oi == 0)
{
UWARN("Cloud with only NaN values created!");
}
UDEBUG("");
return cloud;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloudFromDisparity(
const cv::Mat & imageDisparity,
const StereoCameraModel & model,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
UASSERT(imageDisparity.type() == CV_32FC1 || imageDisparity.type()==CV_16SC1);
UASSERT(decimation >= 1);
if(imageDisparity.rows % decimation != 0 || imageDisparity.cols % decimation != 0)
{
int oldDecimation = decimation;
while(decimation >= 1)
{
if(imageDisparity.rows % decimation == 0 && imageDisparity.cols % decimation == 0)
{
break;
}
--decimation;
}
if(imageDisparity.rows % oldDecimation != 0 || imageDisparity.cols % oldDecimation != 0)
{
UWARN("Decimation (%d) is not valid for current image size (depth=%dx%d). Highest compatible decimation used=%d.", oldDecimation, imageDisparity.cols, imageDisparity.rows, decimation);
}
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
//cloud.header = cameraInfo.header;
cloud->height = imageDisparity.rows/decimation;
cloud->width = imageDisparity.cols/decimation;
cloud->is_dense = false;
cloud->resize(cloud->height * cloud->width);
if(validIndices)
{
validIndices->resize(cloud->size());
}
int oi = 0;
if(imageDisparity.type()==CV_16SC1)
{
for(int h = 0; h < imageDisparity.rows && h/decimation < (int)cloud->height; h+=decimation)
{
for(int w = 0; w < imageDisparity.cols && w/decimation < (int)cloud->width; w+=decimation)
{
float disp = float(imageDisparity.at<short>(h,w))/16.0f;
cv::Point3f pt = projectDisparityTo3D(cv::Point2f(w, h), disp, model);
if(pt.z >= minDepth && (maxDepth <= 0.0f || pt.z <= maxDepth))
{
cloud->at((h/decimation)*cloud->width + (w/decimation)) = pcl::PointXYZ(pt.x, pt.y, pt.z);
if(validIndices)
{
validIndices->at(oi++) = (h/decimation)*cloud->width + (w/decimation);
}
}
else
{
cloud->at((h/decimation)*cloud->width + (w/decimation)) = pcl::PointXYZ(
std::numeric_limits<float>::quiet_NaN(),
std::numeric_limits<float>::quiet_NaN(),
std::numeric_limits<float>::quiet_NaN());
}
}
}
}
else
{
for(int h = 0; h < imageDisparity.rows && h/decimation < (int)cloud->height; h+=decimation)
{
for(int w = 0; w < imageDisparity.cols && w/decimation < (int)cloud->width; w+=decimation)
{
float disp = imageDisparity.at<float>(h,w);
cv::Point3f pt = projectDisparityTo3D(cv::Point2f(w, h), disp, model);
if(pt.z > minDepth && (maxDepth <= 0.0f || pt.z <= maxDepth))
{
cloud->at((h/decimation)*cloud->width + (w/decimation)) = pcl::PointXYZ(pt.x, pt.y, pt.z);
if(validIndices)
{
validIndices->at(oi++) = (h/decimation)*cloud->width + (w/decimation);
}
}
else
{
cloud->at((h/decimation)*cloud->width + (w/decimation)) = pcl::PointXYZ(
std::numeric_limits<float>::quiet_NaN(),
std::numeric_limits<float>::quiet_NaN(),
std::numeric_limits<float>::quiet_NaN());
}
}
}
}
if(validIndices)
{
validIndices->resize(oi);
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudFromDisparityRGB(
const cv::Mat & imageRgb,
const cv::Mat & imageDisparity,
const StereoCameraModel & model,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices)
{
UASSERT(!imageRgb.empty() && !imageDisparity.empty());
UASSERT(imageRgb.rows == imageDisparity.rows &&
imageRgb.cols == imageDisparity.cols &&
(imageDisparity.type() == CV_32FC1 || imageDisparity.type()==CV_16SC1));
UASSERT(imageRgb.channels() == 3 || imageRgb.channels() == 1);
UASSERT(decimation >= 1);
if(imageDisparity.rows % decimation != 0 || imageDisparity.cols % decimation != 0)
{
int oldDecimation = decimation;
while(decimation >= 1)
{
if(imageDisparity.rows % decimation == 0 && imageDisparity.cols % decimation == 0)
{
break;
}
--decimation;
}
if(imageDisparity.rows % oldDecimation != 0 || imageDisparity.cols % oldDecimation != 0)
{
UWARN("Decimation (%d) is not valid for current image size (depth=%dx%d). Highest compatible decimation used=%d.", oldDecimation, imageDisparity.cols, imageDisparity.rows, decimation);
}
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZRGB>);
bool mono;
if(imageRgb.channels() == 3) // BGR
{
mono = false;
}
else // Mono
{
mono = true;
}
//cloud.header = cameraInfo.header;
cloud->height = imageRgb.rows/decimation;
cloud->width = imageRgb.cols/decimation;
cloud->is_dense = false;
cloud->resize(cloud->height * cloud->width);
if(validIndices)
{
validIndices->resize(cloud->size());
}
int oi=0;
for(int h = 0; h < imageRgb.rows && h/decimation < (int)cloud->height; h+=decimation)
{
for(int w = 0; w < imageRgb.cols && w/decimation < (int)cloud->width; w+=decimation)
{
pcl::PointXYZRGB & pt = cloud->at((h/decimation)*cloud->width + (w/decimation));
if(!mono)
{
pt.b = imageRgb.at<cv::Vec3b>(h,w)[0];
pt.g = imageRgb.at<cv::Vec3b>(h,w)[1];
pt.r = imageRgb.at<cv::Vec3b>(h,w)[2];
}
else
{
unsigned char v = imageRgb.at<unsigned char>(h,w);
pt.b = v;
pt.g = v;
pt.r = v;
}
float disp = imageDisparity.type()==CV_16SC1?float(imageDisparity.at<short>(h,w))/16.0f:imageDisparity.at<float>(h,w);
cv::Point3f ptXYZ = projectDisparityTo3D(cv::Point2f(w, h), disp, model);
if(util3d::isFinite(ptXYZ) && ptXYZ.z >= minDepth && (maxDepth<=0.0f || ptXYZ.z <= maxDepth))
{
pt.x = ptXYZ.x;
pt.y = ptXYZ.y;
pt.z = ptXYZ.z;
if(validIndices)
{
validIndices->at(oi++) = (h/decimation)*cloud->width + (w/decimation);
}
}
else
{
pt.x = pt.y = pt.z = std::numeric_limits<float>::quiet_NaN();
}
}
}
if(validIndices)
{
validIndices->resize(oi);
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudFromStereoImages(
const cv::Mat & imageLeft,
const cv::Mat & imageRight,
const StereoCameraModel & model,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices,
const ParametersMap & parameters)
{
UASSERT(!imageLeft.empty() && !imageRight.empty());
UASSERT(imageRight.type() == CV_8UC1 || imageRight.type() == CV_8UC3);
UASSERT(imageLeft.channels() == 3 || imageLeft.channels() == 1);
UASSERT(imageLeft.rows == imageRight.rows &&
imageLeft.cols == imageRight.cols);
UASSERT(decimation >= 1.0f);
cv::Mat leftColor = imageLeft;
cv::Mat rightColor = imageRight;
cv::Mat leftMono;
if(leftColor.channels() == 3)
{
cv::cvtColor(leftColor, leftMono, cv::COLOR_BGR2GRAY);
}
else
{
leftMono = leftColor;
}
cv::Mat rightMono;
if(rightColor.channels() == 3)
{
cv::cvtColor(rightColor, rightMono, cv::COLOR_BGR2GRAY);
}
else
{
rightMono = rightColor;
}
return cloudFromDisparityRGB(
leftColor,
util2d::disparityFromStereoImages(leftMono, rightMono, parameters),
model,
decimation,
maxDepth,
minDepth,
validIndices);
}
std::vector<pcl::PointCloud<pcl::PointXYZ>::Ptr> cloudsFromSensorData(
const SensorData & sensorData,
int decimation,
float maxDepth,
float minDepth,
std::vector<pcl::IndicesPtr> * validIndices,
const ParametersMap & stereoParameters,
const std::vector<float> & roiRatios,
unsigned char confidenceThr)
{
if(decimation == 0)
{
decimation = 1;
}
std::vector<pcl::PointCloud<pcl::PointXYZ>::Ptr> clouds;
if(!sensorData.depthRaw().empty() && sensorData.cameraModels().size())
{
//depth
UASSERT(int((sensorData.depthRaw().cols/sensorData.cameraModels().size())*sensorData.cameraModels().size()) == sensorData.depthRaw().cols);
UASSERT(sensorData.depthConfidenceRaw().empty() || confidenceThr==0 || (sensorData.depthConfidenceRaw().type() == CV_8UC1 && sensorData.depthConfidenceRaw().cols == sensorData.depthRaw().cols && sensorData.depthConfidenceRaw().rows == sensorData.depthRaw().rows));
int subImageWidth = sensorData.depthRaw().cols/sensorData.cameraModels().size();
for(unsigned int i=0; i<sensorData.cameraModels().size(); ++i)
{
clouds.push_back(pcl::PointCloud<pcl::PointXYZ>::Ptr(new pcl::PointCloud<pcl::PointXYZ>));
if(validIndices)
{
validIndices->push_back(pcl::IndicesPtr(new std::vector<int>()));
}
if(sensorData.cameraModels()[i].isValidForProjection())
{
cv::Mat depth = cv::Mat(sensorData.depthRaw(), cv::Rect(subImageWidth*i, 0, subImageWidth, sensorData.depthRaw().rows));
cv::Mat depthConfidence;
if(!sensorData.depthConfidenceRaw().empty() && confidenceThr > 0) {
depthConfidence = cv::Mat(sensorData.depthConfidenceRaw(), cv::Rect(subImageWidth*i, 0, subImageWidth, sensorData.depthConfidenceRaw().rows));
}
CameraModel model = sensorData.cameraModels()[i];
if( roiRatios.size() == 4 &&
(roiRatios[0] > 0.0f ||
roiRatios[1] > 0.0f ||
roiRatios[2] > 0.0f ||
roiRatios[3] > 0.0f))
{
cv::Rect roiDepth = util2d::computeRoi(depth, roiRatios);
cv::Rect roiRgb;
if(model.imageWidth() && model.imageHeight())
{
roiRgb = util2d::computeRoi(model.imageSize(), roiRatios);
}
if( roiDepth.width%decimation==0 &&
roiDepth.height%decimation==0 &&
(roiRgb.width != 0 ||
(roiRgb.width%decimation==0 &&
roiRgb.height%decimation==0)))
{
depth = cv::Mat(depth, roiDepth);
if(!depthConfidence.empty()) {
depthConfidence = cv::Mat(depthConfidence, roiDepth);
}
if(model.imageWidth() != 0 && model.imageHeight() != 0)
{
model = model.roi(util2d::computeRoi(model.imageSize(), roiRatios));
}
else
{
model = model.roi(roiDepth);
}
}
else
{
UERROR("Cannot apply ROI ratios [%f,%f,%f,%f] because resulting "
"dimension (depth=%dx%d rgb=%dx%d) cannot be divided exactly "
"by decimation parameter (%d). Ignoring ROI ratios...",
roiRatios[0],
roiRatios[1],
roiRatios[2],
roiRatios[3],
roiDepth.width,
roiDepth.height,
roiRgb.width,
roiRgb.height,
decimation);
}
}
pcl::PointCloud<pcl::PointXYZ>::Ptr tmp = util3d::cloudFromDepth(
depth,
depthConfidence,
model,
decimation,
maxDepth,
minDepth,
confidenceThr,
validIndices?validIndices->back().get():0);
if(tmp->size())
{
if(!model.localTransform().isNull() && !model.localTransform().isIdentity())
{
tmp = util3d::transformPointCloud(tmp, model.localTransform());
}
clouds.back() = tmp;
}
}
else
{
UERROR("Camera model %d is invalid", i);
}
}
}
else if(!sensorData.imageRaw().empty() && !sensorData.rightRaw().empty() && !sensorData.stereoCameraModels().empty())
{
//stereo
UASSERT(sensorData.rightRaw().type() == CV_8UC1 || sensorData.rightRaw().type() == CV_8UC3);
cv::Mat leftMono;
if(sensorData.imageRaw().channels() == 3)
{
cv::cvtColor(sensorData.imageRaw(), leftMono, cv::COLOR_BGR2GRAY);
}
else
{
leftMono = sensorData.imageRaw();
}
cv::Mat rightMono;
if(sensorData.rightRaw().channels() == 3)
{
cv::cvtColor(sensorData.rightRaw(), rightMono, cv::COLOR_BGR2GRAY);
}
else
{
rightMono = sensorData.rightRaw();
}
UASSERT(int((sensorData.imageRaw().cols/sensorData.stereoCameraModels().size())*sensorData.stereoCameraModels().size()) == sensorData.imageRaw().cols);
UASSERT(int((sensorData.rightRaw().cols/sensorData.stereoCameraModels().size())*sensorData.stereoCameraModels().size()) == sensorData.rightRaw().cols);
int subImageWidth = sensorData.rightRaw().cols/sensorData.stereoCameraModels().size();
for(unsigned int i=0; i<sensorData.stereoCameraModels().size(); ++i)
{
clouds.push_back(pcl::PointCloud<pcl::PointXYZ>::Ptr(new pcl::PointCloud<pcl::PointXYZ>));
if(validIndices)
{
validIndices->push_back(pcl::IndicesPtr(new std::vector<int>()));
}
if(sensorData.stereoCameraModels()[i].isValidForProjection())
{
cv::Mat left(leftMono, cv::Rect(subImageWidth*i, 0, subImageWidth, leftMono.rows));
cv::Mat right(rightMono, cv::Rect(subImageWidth*i, 0, subImageWidth, rightMono.rows));
StereoCameraModel model = sensorData.stereoCameraModels()[i];
if( roiRatios.size() == 4 &&
((roiRatios[0] > 0.0f && roiRatios[0] <= 1.0f) ||
(roiRatios[1] > 0.0f && roiRatios[1] <= 1.0f) ||
(roiRatios[2] > 0.0f && roiRatios[2] <= 1.0f) ||
(roiRatios[3] > 0.0f && roiRatios[3] <= 1.0f)))
{
cv::Rect roi = util2d::computeRoi(left, roiRatios);
if( roi.width%decimation==0 &&
roi.height%decimation==0)
{
left = cv::Mat(left, roi);
right = cv::Mat(right, roi);
model.roi(roi);
}
else
{
UERROR("Cannot apply ROI ratios [%f,%f,%f,%f] because resulting "
"dimension (left=%dx%d) cannot be divided exactly "
"by decimation parameter (%d). Ignoring ROI ratios...",
roiRatios[0],
roiRatios[1],
roiRatios[2],
roiRatios[3],
roi.width,
roi.height,
decimation);
}
}
pcl::PointCloud<pcl::PointXYZ>::Ptr tmp = cloudFromDisparity(
util2d::disparityFromStereoImages(left, right, stereoParameters),
model,
decimation,
maxDepth,
minDepth,
validIndices?validIndices->back().get():0);
if(tmp->size())
{
if(!model.localTransform().isNull() && !model.localTransform().isIdentity())
{
tmp = util3d::transformPointCloud(tmp, model.localTransform());
}
clouds.back() = tmp;
}
}
else
{
UERROR("Stereo camera model %d is invalid", i);
}
}
}
return clouds;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloudFromSensorData(
const SensorData & sensorData,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices,
const ParametersMap & stereoParameters,
const std::vector<float> & roiRatios,
unsigned char confidenceThr)
{
std::vector<pcl::IndicesPtr> validIndicesV;
std::vector<pcl::PointCloud<pcl::PointXYZ>::Ptr> clouds = cloudsFromSensorData(
sensorData,
decimation,
maxDepth,
minDepth,
validIndices?&validIndicesV:0,
stereoParameters,
roiRatios,
confidenceThr);
if(validIndices)
{
UASSERT(validIndicesV.size() == clouds.size());
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
if(clouds.size() == 1)
{
cloud = clouds[0];
if(validIndices)
{
*validIndices = *validIndicesV[0];
}
}
else
{
for(size_t i=0; i<clouds.size(); ++i)
{
*cloud += *util3d::removeNaNFromPointCloud(clouds[i]);
}
if(validIndices)
{
//generate indices for all points (they are all valid)
validIndices->resize(cloud->size());
for(size_t i=0; i<cloud->size(); ++i)
{
validIndices->at(i) = i;
}
}
}
return cloud;
}
std::vector<pcl::PointCloud<pcl::PointXYZRGB>::Ptr> cloudsRGBFromSensorData(
const SensorData & sensorData,
int decimation,
float maxDepth,
float minDepth,
std::vector<pcl::IndicesPtr> * validIndices,
const ParametersMap & stereoParameters,
const std::vector<float> & roiRatios,
unsigned char confidenceThr)
{
if(decimation == 0)
{
decimation = 1;
}
std::vector<pcl::PointCloud<pcl::PointXYZRGB>::Ptr> clouds;
if(!sensorData.imageRaw().empty() && !sensorData.depthRaw().empty() && sensorData.cameraModels().size())
{
//depth
UDEBUG("");
UASSERT(int((sensorData.imageRaw().cols/sensorData.cameraModels().size())*sensorData.cameraModels().size()) == sensorData.imageRaw().cols);
UASSERT(int((sensorData.depthRaw().cols/sensorData.cameraModels().size())*sensorData.cameraModels().size()) == sensorData.depthRaw().cols);
//UASSERT_MSG(sensorData.imageRaw().cols % sensorData.depthRaw().cols == 0, uFormat("rgb=%d depth=%d", sensorData.imageRaw().cols, sensorData.depthRaw().cols).c_str());
//UASSERT_MSG(sensorData.imageRaw().rows % sensorData.depthRaw().rows == 0, uFormat("rgb=%d depth=%d", sensorData.imageRaw().rows, sensorData.depthRaw().rows).c_str());
int subRGBWidth = sensorData.imageRaw().cols/sensorData.cameraModels().size();
int subDepthWidth = sensorData.depthRaw().cols/sensorData.cameraModels().size();
UASSERT(sensorData.depthConfidenceRaw().empty() || confidenceThr==0 || (sensorData.depthConfidenceRaw().type() == CV_8UC1 && sensorData.depthConfidenceRaw().cols == sensorData.depthRaw().cols && sensorData.depthConfidenceRaw().rows == sensorData.depthRaw().rows));
for(unsigned int i=0; i<sensorData.cameraModels().size(); ++i)
{
clouds.push_back(pcl::PointCloud<pcl::PointXYZRGB>::Ptr(new pcl::PointCloud<pcl::PointXYZRGB>));
if(validIndices)
{
validIndices->push_back(pcl::IndicesPtr(new std::vector<int>()));
}
if(sensorData.cameraModels()[i].isValidForProjection())
{
cv::Mat rgb(sensorData.imageRaw(), cv::Rect(subRGBWidth*i, 0, subRGBWidth, sensorData.imageRaw().rows));
cv::Mat depth(sensorData.depthRaw(), cv::Rect(subDepthWidth*i, 0, subDepthWidth, sensorData.depthRaw().rows));
cv::Mat depthConfidence;
if(!sensorData.depthConfidenceRaw().empty() && confidenceThr>0) {
depthConfidence = cv::Mat(sensorData.depthConfidenceRaw(), cv::Rect(subDepthWidth*i, 0, subDepthWidth, sensorData.depthConfidenceRaw().rows));
}
CameraModel model = sensorData.cameraModels()[i];
if( roiRatios.size() == 4 &&
((roiRatios[0] > 0.0f && roiRatios[0] <= 1.0f) ||
(roiRatios[1] > 0.0f && roiRatios[1] <= 1.0f) ||
(roiRatios[2] > 0.0f && roiRatios[2] <= 1.0f) ||
(roiRatios[3] > 0.0f && roiRatios[3] <= 1.0f)))
{
cv::Rect roiDepth = util2d::computeRoi(depth, roiRatios);
cv::Rect roiRgb = util2d::computeRoi(rgb, roiRatios);
if( roiDepth.width%decimation==0 &&
roiDepth.height%decimation==0 &&
roiRgb.width%decimation==0 &&
roiRgb.height%decimation==0)
{
depth = cv::Mat(depth, roiDepth);
if(!depthConfidence.empty()) {
depthConfidence = cv::Mat(depthConfidence, roiDepth);
}
rgb = cv::Mat(rgb, roiRgb);
model = model.roi(roiRgb);
}
else
{
UERROR("Cannot apply ROI ratios [%f,%f,%f,%f] because resulting "
"dimension (depth=%dx%d rgb=%dx%d) cannot be divided exactly "
"by decimation parameter (%d). Ignoring ROI ratios...",
roiRatios[0],
roiRatios[1],
roiRatios[2],
roiRatios[3],
roiDepth.width,
roiDepth.height,
roiRgb.width,
roiRgb.height,
decimation);
}
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr tmp = util3d::cloudFromDepthRGB(
rgb,
depth,
depthConfidence,
model,
decimation,
maxDepth,
minDepth,
confidenceThr,
validIndices?validIndices->back().get():0);
if(tmp->size())
{
if(!model.localTransform().isNull() && !model.localTransform().isIdentity())
{
tmp = util3d::transformPointCloud(tmp, model.localTransform());
}
clouds.back() = tmp;
}
}
else
{
UERROR("Camera model %d is invalid", i);
}
}
}
else if(!sensorData.imageRaw().empty() && !sensorData.rightRaw().empty() && !sensorData.stereoCameraModels().empty())
{
//stereo
UDEBUG("");
UASSERT(int((sensorData.imageRaw().cols/sensorData.stereoCameraModels().size())*sensorData.stereoCameraModels().size()) == sensorData.imageRaw().cols);
UASSERT(int((sensorData.rightRaw().cols/sensorData.stereoCameraModels().size())*sensorData.stereoCameraModels().size()) == sensorData.rightRaw().cols);
int subImageWidth = sensorData.rightRaw().cols/sensorData.stereoCameraModels().size();
for(unsigned int i=0; i<sensorData.stereoCameraModels().size(); ++i)
{
clouds.push_back(pcl::PointCloud<pcl::PointXYZRGB>::Ptr(new pcl::PointCloud<pcl::PointXYZRGB>));
if(validIndices)
{
validIndices->push_back(pcl::IndicesPtr(new std::vector<int>()));
}
if(sensorData.stereoCameraModels()[i].isValidForProjection())
{
cv::Mat left(sensorData.imageRaw(), cv::Rect(subImageWidth*i, 0, subImageWidth, sensorData.imageRaw().rows));
cv::Mat right(sensorData.rightRaw(), cv::Rect(subImageWidth*i, 0, subImageWidth, sensorData.rightRaw().rows));
StereoCameraModel model = sensorData.stereoCameraModels()[i];
if( roiRatios.size() == 4 &&
((roiRatios[0] > 0.0f && roiRatios[0] <= 1.0f) ||
(roiRatios[1] > 0.0f && roiRatios[1] <= 1.0f) ||
(roiRatios[2] > 0.0f && roiRatios[2] <= 1.0f) ||
(roiRatios[3] > 0.0f && roiRatios[3] <= 1.0f)))
{
cv::Rect roi = util2d::computeRoi(left, roiRatios);
if( roi.width%decimation==0 &&
roi.height%decimation==0)
{
left = cv::Mat(left, roi);
right = cv::Mat(right, roi);
model.roi(roi);
}
else
{
UERROR("Cannot apply ROI ratios [%f,%f,%f,%f] because resulting "
"dimension (left=%dx%d) cannot be divided exactly "
"by decimation parameter (%d). Ignoring ROI ratios...",
roiRatios[0],
roiRatios[1],
roiRatios[2],
roiRatios[3],
roi.width,
roi.height,
decimation);
}
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr tmp = cloudFromStereoImages(
left,
right,
model,
decimation,
maxDepth,
minDepth,
validIndices?validIndices->back().get():0,
stereoParameters);
if(tmp->size())
{
if(!model.localTransform().isNull() && !model.localTransform().isIdentity())
{
tmp = util3d::transformPointCloud(tmp, model.localTransform());
}
clouds.back() = tmp;
}
}
else
{
UERROR("Stereo camera model %d is invalid", i);
}
}
}
return clouds;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloudRGBFromSensorData(
const SensorData & sensorData,
int decimation,
float maxDepth,
float minDepth,
std::vector<int> * validIndices,
const ParametersMap & stereoParameters,
const std::vector<float> & roiRatios,
unsigned char confidenceThr)
{
std::vector<pcl::IndicesPtr> validIndicesV;
std::vector<pcl::PointCloud<pcl::PointXYZRGB>::Ptr> clouds = cloudsRGBFromSensorData(
sensorData,
decimation,
maxDepth,
minDepth,
validIndices?&validIndicesV:0,
stereoParameters,
roiRatios,
confidenceThr);
if(validIndices)
{
UASSERT(validIndicesV.size() == clouds.size());
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZRGB>);
if(clouds.size() == 1)
{
cloud = clouds[0];
if(validIndices)
{
*validIndices = *validIndicesV[0];
}
}
else
{
for(size_t i=0; i<clouds.size(); ++i)
{
*cloud += *util3d::removeNaNFromPointCloud(clouds[i]);
}
if(validIndices)
{
//generate indices for all points (they are all valid)
validIndices->resize(cloud->size());
for(size_t i=0; i<cloud->size(); ++i)
{
validIndices->at(i) = i;
}
}
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZ> laserScanFromDepthImage(
const cv::Mat & depthImage,
float fx,
float fy,
float cx,
float cy,
float maxDepth,
float minDepth,
const Transform & localTransform)
{
UASSERT(depthImage.type() == CV_16UC1 || depthImage.type() == CV_32FC1);
UASSERT(!localTransform.isNull());
pcl::PointCloud<pcl::PointXYZ> scan;
int middle = depthImage.rows/2;
if(middle)
{
scan.resize(depthImage.cols);
int oi = 0;
for(int i=depthImage.cols-1; i>=0; --i)
{
pcl::PointXYZ pt = util3d::projectDepthTo3D(depthImage, i, middle, cx, cy, fx, fy, false);
if(pcl::isFinite(pt) && pt.z >= minDepth && (maxDepth == 0 || pt.z < maxDepth))
{
if(!localTransform.isIdentity())
{
pt = util3d::transformPoint(pt, localTransform);
}
scan[oi++] = pt;
}
}
scan.resize(oi);
}
return scan;
}
pcl::PointCloud<pcl::PointXYZ> laserScanFromDepthImages(
const cv::Mat & depthImages,
const std::vector<CameraModel> & cameraModels,
float maxDepth,
float minDepth)
{
pcl::PointCloud<pcl::PointXYZ> scan;
UASSERT(!depthImages.empty() && !cameraModels.empty());
UASSERT(int((depthImages.cols/cameraModels.size())*cameraModels.size()) == depthImages.cols);
int subImageWidth = depthImages.cols/cameraModels.size();
for(int i=(int)cameraModels.size()-1; i>=0; --i)
{
UASSERT(cameraModels[i].isValidForProjection());
UASSERT(cameraModels[i].imageWidth() == subImageWidth);
UASSERT(subImageWidth*(i+1) <= depthImages.cols);
cv::Mat depth = cv::Mat(depthImages, cv::Rect(subImageWidth*i, 0, subImageWidth, depthImages.rows));
scan += laserScanFromDepthImage(
depth,
cameraModels[i].fx(),
cameraModels[i].fy(),
cameraModels[i].cx(),
cameraModels[i].cy(),
maxDepth,
minDepth,
cameraModels[i].localTransform());
}
return scan;
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZ> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZ> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi = 0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC3);
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || pcl::isFinite(cloud.at(index)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZ pt = pcl::transformPoint(cloud.at(index), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
}
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC3);
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZ pt = pcl::transformPoint(cloud.at(i), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
}
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZ);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointNormal> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointNormal> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi=0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC(6));
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || (pcl::isFinite(cloud.at(index)) &&
uIsFinite(cloud.at(index).normal_x) &&
uIsFinite(cloud.at(index).normal_y) &&
uIsFinite(cloud.at(index).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointNormal pt = util3d::transformPoint(cloud.at(index), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
ptr[3] = cloud.at(index).normal_x;
ptr[4] = cloud.at(index).normal_y;
ptr[5] = cloud.at(index).normal_z;
}
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(6));
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) &&
uIsFinite(cloud.at(i).normal_x) &&
uIsFinite(cloud.at(i).normal_y) &&
uIsFinite(cloud.at(i).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointNormal pt = util3d::transformPoint(cloud.at(i), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[3] = cloud.at(i).normal_x;
ptr[4] = cloud.at(i).normal_y;
ptr[5] = cloud.at(i).normal_z;
}
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZNormal);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZ> & cloud, const pcl::PointCloud<pcl::Normal> & normals, const Transform & transform, bool filterNaNs)
{
UASSERT(cloud.size() == normals.size());
cv::Mat laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(6));
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi =0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) && pcl::isFinite(normals.at(i))))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointNormal pt;
pt.x = cloud.at(i).x;
pt.y = cloud.at(i).y;
pt.z = cloud.at(i).z;
pt.normal_x = normals.at(i).normal_x;
pt.normal_y = normals.at(i).normal_y;
pt.normal_z = normals.at(i).normal_z;
pt = util3d::transformPoint(pt, transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[3] = normals.at(i).normal_x;
ptr[4] = normals.at(i).normal_y;
ptr[5] = normals.at(i).normal_z;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZNormal);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZRGB> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZRGB> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi=0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC(4));
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || pcl::isFinite(cloud.at(index)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZRGB pt = pcl::transformPoint(cloud.at(index), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
}
int * ptrInt = (int*)ptr;
ptrInt[3] = int(cloud.at(index).b) | (int(cloud.at(index).g) << 8) | (int(cloud.at(index).r) << 16);
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(4));
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZRGB pt = pcl::transformPoint(cloud.at(i), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
}
int * ptrInt = (int*)ptr;
ptrInt[3] = int(cloud.at(i).b) | (int(cloud.at(i).g) << 8) | (int(cloud.at(i).r) << 16);
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZRGB);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZI> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<rtabmap::PointXYZIRT> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<rtabmap::PointXYZIRT> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
// Layout: [x, y, z, intensity, ring, time] (ring cast to float, values up to
// ~16M are exactly representable so all realistic laser line counts fit).
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi = 0;
const int total = indices.get() ? (int)indices->size() : (int)cloud.size();
laserScan = cv::Mat(1, total, CV_32FC(6));
for(int i=0; i<total; ++i)
{
int index = indices.get() ? indices->at(i) : i;
const rtabmap::PointXYZIRT & src = cloud.at(index);
if(filterNaNs && !pcl::isFinite(src))
{
continue;
}
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZ pt(src.x, src.y, src.z);
pt = pcl::transformPoint(pt, transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = src.x;
ptr[1] = src.y;
ptr[2] = src.z;
}
ptr[3] = src.intensity;
ptr[4] = static_cast<float>(src.ring);
ptr[5] = src.time;
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0, oi)), 0, 0.0f, LaserScan::kXYZIRT);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZI> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi=0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC(4));
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || pcl::isFinite(cloud.at(index)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZI pt = pcl::transformPoint(cloud.at(index), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
}
ptr[3] = cloud.at(index).intensity;
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(4));
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZI pt = pcl::transformPoint(cloud.at(i), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
}
ptr[3] = cloud.at(i).intensity;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZI);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZRGB> & cloud, const pcl::PointCloud<pcl::Normal> & normals, const Transform & transform, bool filterNaNs)
{
UASSERT(cloud.size() == normals.size());
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(7));
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi = 0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZRGBNormal pt;
pt.x = cloud.at(i).x;
pt.y = cloud.at(i).y;
pt.z = cloud.at(i).z;
pt.normal_x = normals.at(i).normal_x;
pt.normal_y = normals.at(i).normal_y;
pt.normal_z = normals.at(i).normal_z;
pt = util3d::transformPoint(pt, transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[4] = normals.at(i).normal_x;
ptr[5] = normals.at(i).normal_y;
ptr[6] = normals.at(i).normal_z;
}
int * ptrInt = (int*)ptr;
ptrInt[3] = int(cloud.at(i).b) | (int(cloud.at(i).g) << 8) | (int(cloud.at(i).r) << 16);
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZRGBNormal);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZRGBNormal> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZRGBNormal> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi = 0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC(7));
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || (pcl::isFinite(cloud.at(index)) &&
uIsFinite(cloud.at(index).normal_x) &&
uIsFinite(cloud.at(index).normal_y) &&
uIsFinite(cloud.at(index).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZRGBNormal pt = util3d::transformPoint(cloud.at(index), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
ptr[4] = cloud.at(index).normal_x;
ptr[5] = cloud.at(index).normal_y;
ptr[6] = cloud.at(index).normal_z;
}
int * ptrInt = (int*)ptr;
ptrInt[3] = int(cloud.at(index).b) | (int(cloud.at(index).g) << 8) | (int(cloud.at(index).r) << 16);
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(7));
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) &&
uIsFinite(cloud.at(i).normal_x) &&
uIsFinite(cloud.at(i).normal_y) &&
uIsFinite(cloud.at(i).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZRGBNormal pt = util3d::transformPoint(cloud.at(i), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[4] = cloud.at(i).normal_x;
ptr[5] = cloud.at(i).normal_y;
ptr[6] = cloud.at(i).normal_z;
}
int * ptrInt = (int*)ptr;
ptrInt[3] = int(cloud.at(i).b) | (int(cloud.at(i).g) << 8) | (int(cloud.at(i).r) << 16);
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZRGBNormal);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZI> & cloud, const pcl::PointCloud<pcl::Normal> & normals, const Transform & transform, bool filterNaNs)
{
UASSERT(cloud.size() == normals.size());
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(7));
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) && pcl::isFinite(normals.at(i))))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZINormal pt;
pt.x = cloud.at(i).x;
pt.y = cloud.at(i).y;
pt.z = cloud.at(i).z;
pt.normal_x = normals.at(i).normal_x;
pt.normal_y = normals.at(i).normal_y;
pt.normal_z = normals.at(i).normal_z;
pt = util3d::transformPoint(pt, transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[4] = normals.at(i).normal_x;
ptr[5] = normals.at(i).normal_y;
ptr[6] = normals.at(i).normal_z;
}
ptr[3] = cloud.at(i).intensity;
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZINormal);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZINormal> & cloud, const Transform & transform, bool filterNaNs)
{
return laserScanFromPointCloud(cloud, pcl::IndicesPtr(), transform, filterNaNs);
}
LaserScan laserScanFromPointCloud(const pcl::PointCloud<pcl::PointXYZINormal> & cloud, const pcl::IndicesPtr & indices, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan;
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi = 0;
if(indices.get())
{
laserScan = cv::Mat(1, (int)indices->size(), CV_32FC(7));
for(unsigned int i=0; i<indices->size(); ++i)
{
int index = indices->at(i);
if(!filterNaNs || (pcl::isFinite(cloud.at(index)) &&
uIsFinite(cloud.at(index).normal_x) &&
uIsFinite(cloud.at(index).normal_y) &&
uIsFinite(cloud.at(index).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZINormal pt = util3d::transformPoint(cloud.at(index), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(index).x;
ptr[1] = cloud.at(index).y;
ptr[2] = cloud.at(index).z;
ptr[4] = cloud.at(index).normal_x;
ptr[5] = cloud.at(index).normal_y;
ptr[6] = cloud.at(index).normal_z;
}
ptr[3] = cloud.at(i).intensity;
}
}
}
else
{
laserScan = cv::Mat(1, (int)cloud.size(), CV_32FC(7));
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) &&
uIsFinite(cloud.at(i).normal_x) &&
uIsFinite(cloud.at(i).normal_y) &&
uIsFinite(cloud.at(i).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZINormal pt = util3d::transformPoint(cloud.at(i), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.z;
ptr[4] = pt.normal_x;
ptr[5] = pt.normal_y;
ptr[6] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).z;
ptr[4] = cloud.at(i).normal_x;
ptr[5] = cloud.at(i).normal_y;
ptr[6] = cloud.at(i).normal_z;
}
ptr[3] = cloud.at(i).intensity;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYZINormal);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointXYZ> & cloud, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC2);
bool nullTransform = transform.isNull();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZ pt = pcl::transformPoint(cloud.at(i), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXY);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointXYZI> & cloud, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC3);
bool nullTransform = transform.isNull();
Eigen::Affine3f transform3f = transform.toEigen3f();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || pcl::isFinite(cloud.at(i)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZI pt = pcl::transformPoint(cloud.at(i), transform3f);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.intensity;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).intensity;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYI);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointNormal> & cloud, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(5));
bool nullTransform = transform.isNull();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) &&
uIsFinite(cloud.at(i).normal_x) &&
uIsFinite(cloud.at(i).normal_y) &&
uIsFinite(cloud.at(i).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointNormal pt = util3d::transformPoint(cloud.at(i), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.normal_x;
ptr[3] = pt.normal_y;
ptr[4] = pt.normal_z;
}
else
{
const pcl::PointNormal & pt = cloud.at(i);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.normal_x;
ptr[3] = pt.normal_y;
ptr[4] = pt.normal_z;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYNormal);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointXYZ> & cloud, const pcl::PointCloud<pcl::Normal> & normals, const Transform & transform, bool filterNaNs)
{
UASSERT(cloud.size() == normals.size());
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(5));
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) && pcl::isFinite(normals.at(i))))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointNormal pt;
pt.x = cloud.at(i).x;
pt.y = cloud.at(i).y;
pt.z = cloud.at(i).z;
pt.normal_x = normals.at(i).normal_x;
pt.normal_y = normals.at(i).normal_y;
pt.normal_z = normals.at(i).normal_z;
pt = util3d::transformPoint(pt, transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.normal_x;
ptr[3] = pt.normal_y;
ptr[4] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = normals.at(i).normal_x;
ptr[3] = normals.at(i).normal_y;
ptr[4] = normals.at(i).normal_z;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYNormal);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointXYZINormal> & cloud, const Transform & transform, bool filterNaNs)
{
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(6));
bool nullTransform = transform.isNull();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) &&
uIsFinite(cloud.at(i).normal_x) &&
uIsFinite(cloud.at(i).normal_y) &&
uIsFinite(cloud.at(i).normal_z)))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZINormal pt = util3d::transformPoint(cloud.at(i), transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.intensity;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
else
{
const pcl::PointXYZINormal & pt = cloud.at(i);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.intensity;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYINormal);
}
LaserScan laserScan2dFromPointCloud(const pcl::PointCloud<pcl::PointXYZI> & cloud, const pcl::PointCloud<pcl::Normal> & normals, const Transform & transform, bool filterNaNs)
{
UASSERT(cloud.size() == normals.size());
cv::Mat laserScan(1, (int)cloud.size(), CV_32FC(6));
bool nullTransform = transform.isNull() || transform.isIdentity();
int oi=0;
for(unsigned int i=0; i<cloud.size(); ++i)
{
if(!filterNaNs || (pcl::isFinite(cloud.at(i)) && pcl::isFinite(normals.at(i))))
{
float * ptr = laserScan.ptr<float>(0, oi++);
if(!nullTransform)
{
pcl::PointXYZINormal pt;
pt.x = cloud.at(i).x;
pt.y = cloud.at(i).y;
pt.z = cloud.at(i).z;
pt.normal_x = normals.at(i).normal_x;
pt.normal_y = normals.at(i).normal_y;
pt.normal_z = normals.at(i).normal_z;
pt = util3d::transformPoint(pt, transform);
ptr[0] = pt.x;
ptr[1] = pt.y;
ptr[2] = pt.intensity;
ptr[3] = pt.normal_x;
ptr[4] = pt.normal_y;
ptr[5] = pt.normal_z;
}
else
{
ptr[0] = cloud.at(i).x;
ptr[1] = cloud.at(i).y;
ptr[2] = cloud.at(i).intensity;
ptr[3] = normals.at(i).normal_x;
ptr[4] = normals.at(i).normal_y;
ptr[5] = normals.at(i).normal_z;
}
}
}
if(oi == 0)
{
return LaserScan();
}
return LaserScan(laserScan(cv::Range::all(), cv::Range(0,oi)), 0, 0.0f, LaserScan::kXYINormal);
}
pcl::PCLPointCloud2::Ptr laserScanToPointCloud2(const LaserScan & laserScan, const Transform & transform)
{
pcl::PCLPointCloud2::Ptr cloud(new pcl::PCLPointCloud2);
if(laserScan.isEmpty())
{
return cloud;
}
if(laserScan.format() == LaserScan::kXY || laserScan.format() == LaserScan::kXYZ)
{
pcl::toPCLPointCloud2(*laserScanToPointCloud(laserScan, transform), *cloud);
}
else if(laserScan.format() == LaserScan::kXYI || laserScan.format() == LaserScan::kXYZI || laserScan.format() == LaserScan::kXYZIT || laserScan.format() == LaserScan::kXYZIRT)
{
pcl::toPCLPointCloud2(*laserScanToPointCloudI(laserScan, transform), *cloud);
}
else if(laserScan.format() == LaserScan::kXYNormal || laserScan.format() == LaserScan::kXYZNormal)
{
pcl::toPCLPointCloud2(*laserScanToPointCloudNormal(laserScan, transform), *cloud);
}
else if(laserScan.format() == LaserScan::kXYINormal || laserScan.format() == LaserScan::kXYZINormal)
{
pcl::toPCLPointCloud2(*laserScanToPointCloudINormal(laserScan, transform), *cloud);
}
else if(laserScan.format() == LaserScan::kXYZRGB)
{
pcl::toPCLPointCloud2(*laserScanToPointCloudRGB(laserScan, transform), *cloud);
}
else if(laserScan.format() == LaserScan::kXYZRGBNormal)
{
pcl::toPCLPointCloud2(*laserScanToPointCloudRGBNormal(laserScan, transform), *cloud);
}
else
{
UERROR("Unknown conversion from LaserScan format %d to PointCloud2.", laserScan.format());
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr laserScanToPointCloud(const LaserScan & laserScan, const Transform & transform)
{
pcl::PointCloud<pcl::PointXYZ>::Ptr output(new pcl::PointCloud<pcl::PointXYZ>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull();
Eigen::Affine3f transform3f = transform.toEigen3f();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = util3d::laserScanToPoint(laserScan, i);
if(!nullTransform)
{
output->at(i) = pcl::transformPoint(output->at(i), transform3f);
}
}
return output;
}
pcl::PointCloud<pcl::PointNormal>::Ptr laserScanToPointCloudNormal(const LaserScan & laserScan, const Transform & transform)
{
pcl::PointCloud<pcl::PointNormal>::Ptr output(new pcl::PointCloud<pcl::PointNormal>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = laserScanToPointNormal(laserScan, i);
if(!nullTransform)
{
output->at(i) = util3d::transformPoint(output->at(i), transform);
}
}
return output;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr laserScanToPointCloudRGB(const LaserScan & laserScan, const Transform & transform, unsigned char r, unsigned char g, unsigned char b)
{
pcl::PointCloud<pcl::PointXYZRGB>::Ptr output(new pcl::PointCloud<pcl::PointXYZRGB>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = util3d::laserScanToPointRGB(laserScan, i, r, g, b);
if(!nullTransform)
{
output->at(i) = pcl::transformPoint(output->at(i), transform3f);
}
}
return output;
}
pcl::PointCloud<pcl::PointXYZI>::Ptr laserScanToPointCloudI(const LaserScan & laserScan, const Transform & transform, float intensity)
{
pcl::PointCloud<pcl::PointXYZI>::Ptr output(new pcl::PointCloud<pcl::PointXYZI>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull() || transform.isIdentity();
Eigen::Affine3f transform3f = transform.toEigen3f();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = util3d::laserScanToPointI(laserScan, i, intensity);
if(!nullTransform)
{
output->at(i) = pcl::transformPoint(output->at(i), transform3f);
}
}
return output;
}
pcl::PointCloud<pcl::PointXYZRGBNormal>::Ptr laserScanToPointCloudRGBNormal(const LaserScan & laserScan, const Transform & transform, unsigned char r, unsigned char g, unsigned char b)
{
pcl::PointCloud<pcl::PointXYZRGBNormal>::Ptr output(new pcl::PointCloud<pcl::PointXYZRGBNormal>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull() || transform.isIdentity();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = util3d::laserScanToPointRGBNormal(laserScan, i, r, g, b);
if(!nullTransform)
{
output->at(i) = util3d::transformPoint(output->at(i), transform);
}
}
return output;
}
pcl::PointCloud<pcl::PointXYZINormal>::Ptr laserScanToPointCloudINormal(const LaserScan & laserScan, const Transform & transform, float intensity)
{
pcl::PointCloud<pcl::PointXYZINormal>::Ptr output(new pcl::PointCloud<pcl::PointXYZINormal>);
if(laserScan.isOrganized())
{
output->width = laserScan.data().cols;
output->height = laserScan.data().rows;
output->is_dense = false;
}
else
{
output->is_dense = true;
}
output->resize(laserScan.size());
bool nullTransform = transform.isNull() || transform.isIdentity();
for(int i=0; i<laserScan.size(); ++i)
{
output->at(i) = util3d::laserScanToPointINormal(laserScan, i, intensity);
if(!nullTransform)
{
output->at(i) = util3d::transformPoint(output->at(i), transform);
}
}
return output;
}
pcl::PointXYZ laserScanToPoint(const LaserScan & laserScan, int index)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointXYZ output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
return output;
}
pcl::PointNormal laserScanToPointNormal(const LaserScan & laserScan, int index)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointNormal output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
if(laserScan.hasNormals())
{
int offset = laserScan.getNormalsOffset();
output.normal_x = ptr[offset];
output.normal_y = ptr[offset+1];
output.normal_z = ptr[offset+2];
}
return output;
}
pcl::PointXYZRGB laserScanToPointRGB(const LaserScan & laserScan, int index, unsigned char r, unsigned char g, unsigned char b)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointXYZRGB output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
if(laserScan.hasRGB())
{
LaserScan::unpackRGB(ptr[laserScan.getRGBOffset()], output.r, output.g, output.b);
}
else if(laserScan.hasIntensity())
{
// package intensity float -> rgba
int * ptrInt = (int*)ptr;
int indexIntensity = laserScan.getIntensityOffset();
output.r = (unsigned char)(ptrInt[indexIntensity] & 0xFF);
output.g = (unsigned char)((ptrInt[indexIntensity] >> 8) & 0xFF);
output.b = (unsigned char)((ptrInt[indexIntensity] >> 16) & 0xFF);
output.a = (unsigned char)((ptrInt[indexIntensity] >> 24) & 0xFF);
}
else
{
output.r = r;
output.g = g;
output.b = b;
}
return output;
}
pcl::PointXYZI laserScanToPointI(const LaserScan & laserScan, int index, float intensity)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointXYZI output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
if(laserScan.hasIntensity())
{
int offset = laserScan.getIntensityOffset();
output.intensity = ptr[offset];
}
else
{
output.intensity = intensity;
}
return output;
}
pcl::PointXYZRGBNormal laserScanToPointRGBNormal(const LaserScan & laserScan, int index, unsigned char r, unsigned char g, unsigned char b)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointXYZRGBNormal output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
if(laserScan.hasRGB())
{
LaserScan::unpackRGB(ptr[laserScan.getRGBOffset()], output.r, output.g, output.b);
}
else if(laserScan.hasIntensity())
{
int * ptrInt = (int*)ptr;
int indexIntensity = laserScan.getIntensityOffset();
output.r = (unsigned char)(ptrInt[indexIntensity] & 0xFF);
output.g = (unsigned char)((ptrInt[indexIntensity] >> 8) & 0xFF);
output.b = (unsigned char)((ptrInt[indexIntensity] >> 16) & 0xFF);
output.a = (unsigned char)((ptrInt[indexIntensity] >> 24) & 0xFF);
}
else
{
output.r = r;
output.g = g;
output.b = b;
}
if(laserScan.hasNormals())
{
int offset = laserScan.getNormalsOffset();
output.normal_x = ptr[offset];
output.normal_y = ptr[offset+1];
output.normal_z = ptr[offset+2];
}
return output;
}
pcl::PointXYZINormal laserScanToPointINormal(const LaserScan & laserScan, int index, float intensity)
{
UASSERT(!laserScan.isEmpty() && !laserScan.isCompressed() && index < laserScan.size());
pcl::PointXYZINormal output;
int row = index / laserScan.data().cols;
const float * ptr = laserScan.data().ptr<float>(row, index - row*laserScan.data().cols);
output.x = ptr[0];
output.y = ptr[1];
if(!laserScan.is2d())
{
output.z = ptr[2];
}
if(laserScan.hasIntensity())
{
int offset = laserScan.getIntensityOffset();
output.intensity = ptr[offset];
}
else
{
output.intensity = intensity;
}
if(laserScan.hasNormals())
{
int offset = laserScan.getNormalsOffset();
output.normal_x = ptr[offset];
output.normal_y = ptr[offset+1];
output.normal_z = ptr[offset+2];
}
return output;
}
void getMinMax3D(const cv::Mat & laserScan, cv::Point3f & min, cv::Point3f & max)
{
UASSERT(!laserScan.empty());
UASSERT(laserScan.type() == CV_32FC2 || laserScan.type() == CV_32FC3 || laserScan.type() == CV_32FC(4) || laserScan.type() == CV_32FC(5) || laserScan.type() == CV_32FC(6) || laserScan.type() == CV_32FC(7));
const float * ptr = laserScan.ptr<float>(0, 0);
min.x = max.x = ptr[0];
min.y = max.y = ptr[1];
bool is3d = laserScan.channels() >= 3 && laserScan.channels() != 5;
min.z = max.z = is3d?ptr[2]:0.0f;
for(int i=1; i<laserScan.cols; ++i)
{
ptr = laserScan.ptr<float>(0, i);
if(ptr[0] < min.x) min.x = ptr[0];
else if(ptr[0] > max.x) max.x = ptr[0];
if(ptr[1] < min.y) min.y = ptr[1];
else if(ptr[1] > max.y) max.y = ptr[1];
if(is3d)
{
if(ptr[2] < min.z) min.z = ptr[2];
else if(ptr[2] > max.z) max.z = ptr[2];
}
}
}
void getMinMax3D(const cv::Mat & laserScan, pcl::PointXYZ & min, pcl::PointXYZ & max)
{
cv::Point3f minCV, maxCV;
getMinMax3D(laserScan, minCV, maxCV);
min.x = minCV.x;
min.y = minCV.y;
min.z = minCV.z;
max.x = maxCV.x;
max.y = maxCV.y;
max.z = maxCV.z;
}
// inspired from ROS image_geometry/src/stereo_camera_model.cpp
cv::Point3f projectDisparityTo3D(
const cv::Point2f & pt,
float disparity,
const StereoCameraModel & model)
{
if(disparity > 0.0f && model.baseline() > 0.0f && model.left().fx() > 0.0f)
{
//Z = baseline * f / (d + cx1-cx0);
float c = 0.0f;
if(model.right().cx()>0.0f && model.left().cx()>0.0f)
{
c = model.right().cx() - model.left().cx();
}
float W = model.baseline()/(disparity + c);
return cv::Point3f((pt.x - model.left().cx())*W,
(pt.y - model.left().cy())*model.left().fx()/model.left().fy()*W, model.left().fx()*W);
}
float bad_point = std::numeric_limits<float>::quiet_NaN ();
return cv::Point3f(bad_point, bad_point, bad_point);
}
cv::Point3f projectDisparityTo3D(
const cv::Point2f & pt,
const cv::Mat & disparity,
const StereoCameraModel & model)
{
UASSERT(!disparity.empty() && (disparity.type() == CV_32FC1 || disparity.type() == CV_16SC1));
int u = int(pt.x+0.5f);
int v = int(pt.y+0.5f);
float bad_point = std::numeric_limits<float>::quiet_NaN ();
if(uIsInBounds(u, 0, disparity.cols) &&
uIsInBounds(v, 0, disparity.rows))
{
float d = disparity.type() == CV_16SC1?float(disparity.at<short>(v,u))/16.0f:disparity.at<float>(v,u);
return projectDisparityTo3D(pt, d, model);
}
return cv::Point3f(bad_point, bad_point, bad_point);
}
// Register point cloud to camera (return registered depth image)
cv::Mat projectCloudToCamera(
const cv::Size & imageSize,
const cv::Mat & cameraMatrixK,
const cv::Mat & laserScan, // assuming laser scan points are already in /base_link coordinate
const rtabmap::Transform & cameraTransform) // /base_link -> /camera_link
{
UASSERT(!cameraTransform.isNull());
UASSERT(!laserScan.empty());
UASSERT(laserScan.type() == CV_32FC2 || laserScan.type() == CV_32FC3 || laserScan.type() == CV_32FC(4) || laserScan.type() == CV_32FC(5) || laserScan.type() == CV_32FC(6) || laserScan.type() == CV_32FC(7));
UASSERT(cameraMatrixK.type() == CV_64FC1 && cameraMatrixK.cols == 3 && cameraMatrixK.cols == 3);
float fx = cameraMatrixK.at<double>(0,0);
float fy = cameraMatrixK.at<double>(1,1);
float cx = cameraMatrixK.at<double>(0,2);
float cy = cameraMatrixK.at<double>(1,2);
cv::Mat registered = cv::Mat::zeros(imageSize, CV_32FC1);
Transform t = cameraTransform.inverse();
int count = 0;
for(int i=0; i<laserScan.cols; ++i)
{
const float* ptr = laserScan.ptr<float>(0, i);
// Get 3D from laser scan
cv::Point3f ptScan;
if(laserScan.type() == CV_32FC2 || laserScan.type() == CV_32FC(5))
{
// 2D scans
ptScan.x = ptr[0];
ptScan.y = ptr[1];
ptScan.z = 0;
}
else // 3D scans
{
ptScan.x = ptr[0];
ptScan.y = ptr[1];
ptScan.z = ptr[2];
}
ptScan = util3d::transformPoint(ptScan, t);
// re-project in camera frame
float z = ptScan.z;
bool set = false;
if(z > 0.0f)
{
float invZ = 1.0f/z;
float dx = (fx*ptScan.x)*invZ + cx;
float dy = (fy*ptScan.y)*invZ + cy;
int dx_low = dx;
int dy_low = dy;
int dx_high = dx + 0.5f;
int dy_high = dy + 0.5f;
if(uIsInBounds(dx_low, 0, registered.cols) && uIsInBounds(dy_low, 0, registered.rows))
{
float &zReg = registered.at<float>(dy_low, dx_low);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
set = true;
}
if((dx_low != dx_high || dy_low != dy_high) &&
uIsInBounds(dx_high, 0, registered.cols) && uIsInBounds(dy_high, 0, registered.rows))
{
float &zReg = registered.at<float>(dy_high, dx_high);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
set = true;
}
}
if(set)
{
count++;
}
}
UDEBUG("Points in camera=%d/%d", count, laserScan.cols);
return registered;
}
cv::Mat projectCloudToCamera(
const cv::Size & imageSize,
const cv::Mat & cameraMatrixK,
const pcl::PointCloud<pcl::PointXYZ>::Ptr laserScan, // assuming points are already in /base_link coordinate
const rtabmap::Transform & cameraTransform) // /base_link -> /camera_link
{
UASSERT(!cameraTransform.isNull());
UASSERT(!laserScan->empty());
UASSERT(cameraMatrixK.type() == CV_64FC1 && cameraMatrixK.cols == 3 && cameraMatrixK.cols == 3);
float fx = cameraMatrixK.at<double>(0,0);
float fy = cameraMatrixK.at<double>(1,1);
float cx = cameraMatrixK.at<double>(0,2);
float cy = cameraMatrixK.at<double>(1,2);
cv::Mat registered = cv::Mat::zeros(imageSize, CV_32FC1);
Transform t = cameraTransform.inverse();
int count = 0;
for(int i=0; i<(int)laserScan->size(); ++i)
{
// Get 3D from laser scan
pcl::PointXYZ ptScan = laserScan->at(i);
ptScan = util3d::transformPoint(ptScan, t);
// re-project in camera frame
float z = ptScan.z;
bool set = false;
if(z > 0.0f)
{
float invZ = 1.0f/z;
float dx = (fx*ptScan.x)*invZ + cx;
float dy = (fy*ptScan.y)*invZ + cy;
int dx_low = dx;
int dy_low = dy;
int dx_high = dx + 0.5f;
int dy_high = dy + 0.5f;
if(uIsInBounds(dx_low, 0, registered.cols) && uIsInBounds(dy_low, 0, registered.rows))
{
set = true;
float &zReg = registered.at<float>(dy_low, dx_low);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
}
if((dx_low != dx_high || dy_low != dy_high) &&
uIsInBounds(dx_high, 0, registered.cols) && uIsInBounds(dy_high, 0, registered.rows))
{
set = true;
float &zReg = registered.at<float>(dy_high, dx_high);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
}
}
if(set)
{
count++;
}
}
UDEBUG("Points in camera=%d/%d", count, (int)laserScan->size());
return registered;
}
cv::Mat projectCloudToCamera(
const cv::Size & imageSize,
const cv::Mat & cameraMatrixK,
const pcl::PCLPointCloud2::Ptr laserScan, // assuming points are already in /base_link coordinate
const rtabmap::Transform & cameraTransform) // /base_link -> /camera_link
{
UASSERT(!cameraTransform.isNull());
UASSERT(!laserScan->data.empty());
UASSERT(cameraMatrixK.type() == CV_64FC1 && cameraMatrixK.cols == 3 && cameraMatrixK.cols == 3);
float fx = cameraMatrixK.at<double>(0,0);
float fy = cameraMatrixK.at<double>(1,1);
float cx = cameraMatrixK.at<double>(0,2);
float cy = cameraMatrixK.at<double>(1,2);
cv::Mat registered = cv::Mat::zeros(imageSize, CV_32FC1);
Transform t = cameraTransform.inverse();
pcl::MsgFieldMap field_map;
pcl::createMapping<pcl::PointXYZ> (laserScan->fields, field_map);
int count = 0;
if(field_map.size() == 1)
{
for (uint32_t row = 0; row < (uint32_t)laserScan->height; ++row)
{
const uint8_t* row_data = &laserScan->data[row * laserScan->row_step];
for (uint32_t col = 0; col < (uint32_t)laserScan->width; ++col)
{
const uint8_t* msg_data = row_data + col * laserScan->point_step;
pcl::PointXYZ ptScan;
memcpy (&ptScan, msg_data + field_map.front().serialized_offset, field_map.front().size);
ptScan = util3d::transformPoint(ptScan, t);
// re-project in camera frame
float z = ptScan.z;
bool set = false;
if(z > 0.0f)
{
float invZ = 1.0f/z;
float dx = (fx*ptScan.x)*invZ + cx;
float dy = (fy*ptScan.y)*invZ + cy;
int dx_low = dx;
int dy_low = dy;
int dx_high = dx + 0.5f;
int dy_high = dy + 0.5f;
if(uIsInBounds(dx_low, 0, registered.cols) && uIsInBounds(dy_low, 0, registered.rows))
{
set = true;
float &zReg = registered.at<float>(dy_low, dx_low);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
}
if((dx_low != dx_high || dy_low != dy_high) &&
uIsInBounds(dx_high, 0, registered.cols) && uIsInBounds(dy_high, 0, registered.rows))
{
set = true;
float &zReg = registered.at<float>(dy_high, dx_high);
if(zReg == 0 || z < zReg)
{
zReg = z;
}
}
}
if(set)
{
count++;
}
}
}
}
else
{
UERROR("field map pcl::pointXYZ not found!");
}
UDEBUG("Points in camera=%d/%d", count, (int)laserScan->data.size());
return registered;
}
void fillProjectedCloudHoles(cv::Mat & registeredDepth, bool verticalDirection, bool fillToBorder)
{
UASSERT(registeredDepth.type() == CV_32FC1);
if(verticalDirection)
{
// vertical, for each column
for(int x=0; x<registeredDepth.cols; ++x)
{
float valueA = 0.0f;
int indexA = -1;
for(int y=0; y<registeredDepth.rows; ++y)
{
float v = registeredDepth.at<float>(y,x);
if(fillToBorder && y == registeredDepth.rows-1 && v<=0.0f && indexA>=0)
{
v = valueA;
}
if(v > 0.0f)
{
if(fillToBorder && indexA < 0)
{
indexA = 0;
valueA = v;
}
if(indexA >=0)
{
int range = y-indexA;
if(range > 1)
{
float slope = (v-valueA)/(range);
for(int k=1; k<range; ++k)
{
registeredDepth.at<float>(indexA+k,x) = valueA+slope*float(k);
}
}
}
valueA = v;
indexA = y;
}
}
}
}
else
{
// horizontal, for each row
for(int y=0; y<registeredDepth.rows; ++y)
{
float valueA = 0.0f;
int indexA = -1;
for(int x=0; x<registeredDepth.cols; ++x)
{
float v = registeredDepth.at<float>(y,x);
if(fillToBorder && x == registeredDepth.cols-1 && v<=0.0f && indexA>=0)
{
v = valueA;
}
if(v > 0.0f)
{
if(fillToBorder && indexA < 0)
{
indexA = 0;
valueA = v;
}
if(indexA >=0)
{
int range = x-indexA;
if(range > 1)
{
float slope = (v-valueA)/(range);
for(int k=1; k<range; ++k)
{
registeredDepth.at<float>(y,indexA+k) = valueA+slope*float(k);
}
}
}
valueA = v;
indexA = x;
}
}
}
}
}
cv::Mat filterFloor(const cv::Mat & depth, const std::vector<CameraModel> & cameraModels, float threshold, cv::Mat * depthBelow)
{
cv::Mat output = depth.clone();
if(depth.empty())
{
return output;
}
if(depthBelow)
{
*depthBelow = cv::Mat::zeros(output.size(), output.type());
}
UASSERT(!cameraModels.empty());
UASSERT(cameraModels[0].isValidForReprojection());
// Support camera model with different resolution than depth image
float rgbToDepthFactorX = float(cameraModels[0].imageWidth()) / float(output.cols/cameraModels.size());
float rgbToDepthFactorY = float(cameraModels[0].imageHeight()) / float(output.rows);
int depthWidth = output.cols/cameraModels.size();
UASSERT(depthWidth*(int)cameraModels.size() == output.cols);
// for each camera
for(size_t i=0; i<cameraModels.size(); ++i)
{
const CameraModel & cam = cameraModels[i];
UASSERT(cam.isValidForReprojection());
const Transform & localTransform = cam.localTransform();
UASSERT(!localTransform.isNull());
if(i>0)
{
// Make sure all models are the same resolution
UASSERT(cam.imageWidth() == cameraModels[i-1].imageWidth());
UASSERT(cam.imageHeight() == cameraModels[i-1].imageHeight());
}
float depthFx = cam.fx() / rgbToDepthFactorX;
float depthFy = cam.fy() / rgbToDepthFactorY;
float depthCx = cam.cx() / rgbToDepthFactorX;
float depthCy = cam.cy() / rgbToDepthFactorY;
cv::Mat subImage = output.colRange(cv::Range(i*depthWidth, (i+1)*depthWidth));
cv::Mat subImageBelow;
if(depthBelow)
subImageBelow = depthBelow->colRange(cv::Range(i*depthWidth, (i+1)*depthWidth));
for(int y=0; y<subImage.rows; ++y)
{
if(subImage.type() == CV_16UC1)
{
unsigned short * ptr = (unsigned short *)subImage.row(y).ptr();
unsigned short * ptrBelow = 0;
if(depthBelow)
{
ptrBelow = (unsigned short *)subImageBelow.row(y).ptr();
}
for(int x=0; x<subImage.cols; ++x)
{
if(ptr[x] > 0)
{
float d = float(ptr[x])/1000.0f;
cv::Point3f pt;
pt.x = (x - depthCx) * d / depthFx;
pt.y = (y - depthCy) * d / depthFy;
pt.z = d;
pt = util3d::transformPoint(pt, localTransform);
if(pt.z < threshold)
{
if(ptrBelow)
{
ptrBelow[x] = ptr[x];
}
ptr[x] = 0;
}
}
}
}
else // CV_32FC1
{
float * ptr = (float *)subImage.row(y).ptr();
float * ptrBelow = 0;
if(depthBelow)
{
ptrBelow = (float *)subImageBelow.row(y).ptr();
}
for(int x=0; x<subImage.cols; ++x)
{
if(ptr[x] > 0.0f)
{
float & d = ptr[x];
cv::Point3f pt;
pt.x = (x - depthCx) * d / depthFx;
pt.y = (y - depthCy) * d / depthFy;
pt.z = d;
pt = util3d::transformPoint(pt, localTransform);
if(pt.z < threshold)
{
if(ptrBelow)
{
ptrBelow[x] = ptr[x];
}
d = 0;
}
}
}
}
}
}
return output;
}
class ProjectionInfo {
public:
ProjectionInfo():
nodeID(-1),
cameraIndex(-1),
distance(-1)
{}
int nodeID;
int cameraIndex;
pcl::PointXY uv;
float distance;
};
class RegisteredPoints {
public:
class Point {
public:
Point(float distance_, int index_) : distance(distance_), index(index_) {}
float distance;
int index;
};
float minDistance;
std::vector<Point> points;
};
/**
* For each point, return pixel of the best camera (NodeID->CameraIndex)
* looking at it based on the policy and parameters
*/
template<class PointT>
std::vector<std::pair< std::pair<int, int>, pcl::PointXY> > projectCloudToCamerasImpl (
const typename pcl::PointCloud<PointT> & cloud,
const std::map<int, Transform> & cameraPoses,
const std::map<int, std::vector<CameraModel> > & cameraModels,
float maxDistance,
float maxAngle,
float maxDepthError,
const std::vector<float> & roiRatios,
const cv::Mat & projMask,
bool distanceToCamPolicy,
const ProgressState * state)
{
UINFO("cloud=%d points", (int)cloud.size());
UINFO("cameraPoses=%d", (int)cameraPoses.size());
UINFO("cameraModels=%d", (int)cameraModels.size());
UINFO("maxDistance=%f", maxDistance);
UINFO("maxAngle=%f", maxAngle);
UINFO("maxDepthError=%f", maxDepthError);
UINFO("distanceToCamPolicy=%s", distanceToCamPolicy?"true":"false");
UINFO("roiRatios=%s", roiRatios.size() == 4?uFormat("%f %f %f %f", roiRatios[0], roiRatios[1], roiRatios[2], roiRatios[3]).c_str():"");
UINFO("projMask=%dx%d", projMask.cols, projMask.rows);
std::vector<std::pair< std::pair<int, int>, pcl::PointXY> > pointToPixel;
if (cloud.empty() || cameraPoses.empty() || cameraModels.empty())
return pointToPixel;
std::string msg = uFormat("Computing visible points per cam (%d points, %d cams)", (int)cloud.size(), (int)cameraPoses.size());
UINFO("%s", msg.c_str());
if(state && !state->callback(msg))
{
//cancelled!
UWARN("Projecting to cameras cancelled!");
return pointToPixel;
}
std::vector<ProjectionInfo> invertedIndex(cloud.size()); // For each point: list of cameras
int cameraProcessed = 0;
bool wrongMaskFormatWarned = false;
for(std::map<int, Transform>::const_iterator pter = cameraPoses.lower_bound(0); pter!=cameraPoses.end(); ++pter)
{
std::map<int, std::vector<CameraModel> >::const_iterator iter=cameraModels.find(pter->first);
if(iter!=cameraModels.end() && !iter->second.empty())
{
cv::Mat validProjMask;
if(!projMask.empty())
{
if(projMask.type() != CV_8UC1)
{
if(!wrongMaskFormatWarned)
UERROR("Wrong camera projection mask type %d, should be CV_8UC1", projMask.type());
wrongMaskFormatWarned = true;
}
else if(projMask.cols == iter->second[0].imageWidth() * (int)iter->second.size() &&
projMask.rows == iter->second[0].imageHeight())
{
validProjMask = projMask;
}
else
{
UWARN("Camera projection mask (%dx%d) is not valid for current "
"camera model(s) (count=%ld, image size=%dx%d). It will be "
"ignored for node %d",
projMask.cols, projMask.rows,
iter->second.size(),
iter->second[0].imageWidth(),
iter->second[0].imageHeight(),
pter->first);
}
}
for(size_t camIndex=0; camIndex<iter->second.size(); ++camIndex)
{
Transform cameraTransform = (pter->second * iter->second[camIndex].localTransform());
UASSERT(!cameraTransform.isNull());
cv::Mat cameraMatrixK = iter->second[camIndex].K();
UASSERT(cameraMatrixK.type() == CV_64FC1 && cameraMatrixK.cols == 3 && cameraMatrixK.cols == 3);
const cv::Size & imageSize = iter->second[camIndex].imageSize();
float fx = cameraMatrixK.at<double>(0,0);
float fy = cameraMatrixK.at<double>(1,1);
float cx = cameraMatrixK.at<double>(0,2);
float cy = cameraMatrixK.at<double>(1,2);
// [rows][cols][depth, indexPt]
std::vector<std::vector<RegisteredPoints> > registered(
imageSize.height, std::vector<RegisteredPoints>(imageSize.width));
Transform t = cameraTransform.inverse();
cv::Rect roi(0,0,imageSize.width, imageSize.height);
if(roiRatios.size()==4)
{
roi = util2d::computeRoi(imageSize, roiRatios);
}
int count = 0;
for(size_t i=0; i<cloud.size(); ++i)
{
// Get 3D from laser scan
PointT ptScan = cloud.at(i);
ptScan = util3d::transformPoint(ptScan, t);
// re-project in camera frame
float z = ptScan.z;
bool set = false;
if(z > 0.0f && (maxDistance<=0 || z<maxDistance))
{
float invZ = 1.0f/z;
float u = (fx*ptScan.x)*invZ + cx;
float v = (fy*ptScan.y)*invZ + cy;
int x = u + 0.5f;
int y = v + 0.5f;
if(uIsInBounds(x, roi.x, roi.x+roi.width) && uIsInBounds(y, roi.y, roi.y+roi.height) &&
(validProjMask.empty() || validProjMask.at<unsigned char>(y, imageSize.width*camIndex+x) > 0)) {
RegisteredPoints &zReg = registered[y][x];
if(zReg.points.empty()) {
zReg.minDistance = z;
zReg.points.push_back(RegisteredPoints::Point(z, i));
set = true;
}
else if(z < zReg.minDistance) {
zReg.minDistance = z;
if(maxDepthError<=0.0f) {
// keeping only closest point, just update it
zReg.points[0].distance = z;
zReg.points[0].index = i;
}
else {
// update the points attached to same pixel based on new closest distance
std::vector<RegisteredPoints::Point> reOrderedPts;
reOrderedPts.push_back(RegisteredPoints::Point(z, i));
for(size_t p=0; p<zReg.points.size(); ++p) {
if(zReg.points[p].distance - z < maxDepthError) {
reOrderedPts.push_back(zReg.points[p]);
}
}
zReg.points = reOrderedPts;
}
set = true;
}
else if(maxDepthError>=0.0f && z - zReg.minDistance < maxDepthError) {
// The point is closer than current closest one to camera,
// but still under max depth difference, just append
zReg.points.push_back(RegisteredPoints::Point(z, i));
set = true;
}
}
}
if(set)
{
count++;
}
}
if(count == 0)
{
registered.clear();
UINFO("No points projected in camera %d/%d", pter->first, (int)camIndex);
}
else
{
UDEBUG("%d points projected in camera %d/%d", count, pter->first, (int)camIndex);
}
for(int u=0; u<imageSize.width; ++u)
{
for(int v=0; v<imageSize.height; ++v)
{
RegisteredPoints &zReg = registered[v][u];
if(!zReg.points.empty())
{
ProjectionInfo info;
info.nodeID = pter->first;
info.cameraIndex = camIndex;
info.uv.x = float(u)/float(imageSize.width);
info.uv.y = float(v)/float(imageSize.height);
const Transform & cam = cameraPoses.at(info.nodeID);
for(size_t p=0; p<zReg.points.size(); ++p)
{
int ptIdx = zReg.points[p].index;
const PointT & pt = cloud.at(ptIdx);
Eigen::Vector4f camDir(cam.x()-pt.x, cam.y()-pt.y, cam.z()-pt.z, 0);
Eigen::Vector4f normal(pt.normal_x, pt.normal_y, pt.normal_z, 0);
float angleToCam = maxAngle<=0?0:pcl::getAngle3D(normal, camDir);
float distanceToCam = zReg.points[p].distance;
if( (maxAngle<=0 || (camDir.dot(normal) > 0 && angleToCam < maxAngle)) && // is facing camera? is point normal perpendicular to camera?
(maxDistance<=0 || distanceToCam<maxDistance)) // is point not too far from camera?
{
float vx = info.uv.x-0.5f;
float vy = info.uv.y-0.5f;
float distanceToCenter = vx*vx+vy*vy;
float distance = distanceToCenter;
if(distanceToCamPolicy)
{
distance = distanceToCam;
}
info.distance = distance;
if(invertedIndex[ptIdx].distance != -1.0f)
{
if(distance <= invertedIndex[ptIdx].distance)
{
invertedIndex[ptIdx] = info;
}
}
else
{
invertedIndex[ptIdx] = info;
}
}
}
}
}
}
}
}
msg = uFormat("Processed camera %d/%d", (int)cameraProcessed+1, (int)cameraPoses.size());
UINFO("%s", msg.c_str());
if(state && !state->callback(msg))
{
//cancelled!
UWARN("Projecting to cameras cancelled!");
return pointToPixel;
}
++cameraProcessed;
}
msg = uFormat("Select best camera for %d points...", (int)cloud.size());
UINFO("%s", msg.c_str());
if(state && !state->callback(msg))
{
//cancelled!
UWARN("Projecting to cameras cancelled!");
return pointToPixel;
}
pointToPixel.resize(invertedIndex.size());
int colorized = 0;
// For each point
for(size_t i=0; i<invertedIndex.size(); ++i)
{
int nodeID = -1;
int cameraIndex = -1;
pcl::PointXY uv_coords;
if(invertedIndex[i].distance > -1.0f)
{
nodeID = invertedIndex[i].nodeID;
cameraIndex = invertedIndex[i].cameraIndex;
uv_coords = invertedIndex[i].uv;
}
if(nodeID>-1 && cameraIndex> -1)
{
pointToPixel[i].first.first = nodeID;
pointToPixel[i].first.second = cameraIndex;
pointToPixel[i].second = uv_coords;
++colorized;
}
}
msg = uFormat("Process %d points...done! (%d [%d%%] projected in cameras)", (int)cloud.size(), colorized, (int)(colorized*100/cloud.size()));
UINFO("%s", msg.c_str());
if(state)
{
state->callback(msg);
}
return pointToPixel;
}
std::vector<std::pair< std::pair<int, int>, pcl::PointXY> > projectCloudToCameras (
const typename pcl::PointCloud<pcl::PointXYZRGBNormal> & cloud,
const std::map<int, Transform> & cameraPoses,
const std::map<int, std::vector<CameraModel> > & cameraModels,
float maxDistance,
float maxAngle,
float maxDepthError,
const std::vector<float> & roiRatios,
const cv::Mat & projMask,
bool distanceToCamPolicy,
const ProgressState * state)
{
return projectCloudToCamerasImpl(cloud,
cameraPoses,
cameraModels,
maxDistance,
maxAngle,
maxDepthError,
roiRatios,
projMask,
distanceToCamPolicy,
state);
}
std::vector<std::pair< std::pair<int, int>, pcl::PointXY> > projectCloudToCameras (
const typename pcl::PointCloud<pcl::PointXYZINormal> & cloud,
const std::map<int, Transform> & cameraPoses,
const std::map<int, std::vector<CameraModel> > & cameraModels,
float maxDistance,
float maxAngle,
float maxDepthError,
const std::vector<float> & roiRatios,
const cv::Mat & projMask,
bool distanceToCamPolicy,
const ProgressState * state)
{
return projectCloudToCamerasImpl(cloud,
cameraPoses,
cameraModels,
maxDistance,
maxAngle,
maxDepthError,
roiRatios,
projMask,
distanceToCamPolicy,
state);
}
bool isFinite(const cv::Point3f & pt)
{
return uIsFinite(pt.x) && uIsFinite(pt.y) && uIsFinite(pt.z);
}
pcl::PointCloud<pcl::PointXYZ>::Ptr concatenateClouds(const std::list<pcl::PointCloud<pcl::PointXYZ>::Ptr> & clouds)
{
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
for(std::list<pcl::PointCloud<pcl::PointXYZ>::Ptr>::const_iterator iter = clouds.begin(); iter!=clouds.end(); ++iter)
{
*cloud += *(*iter);
}
return cloud;
}
pcl::PointCloud<pcl::PointXYZRGB>::Ptr concatenateClouds(const std::list<pcl::PointCloud<pcl::PointXYZRGB>::Ptr> & clouds)
{
pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZRGB>);
for(std::list<pcl::PointCloud<pcl::PointXYZRGB>::Ptr>::const_iterator iter = clouds.begin(); iter!=clouds.end(); ++iter)
{
*cloud+=*(*iter);
}
return cloud;
}
pcl::IndicesPtr concatenate(const std::vector<pcl::IndicesPtr> & indices)
{
//compute total size
unsigned int totalSize = 0;
for(unsigned int i=0; i<indices.size(); ++i)
{
totalSize += (unsigned int)indices[i]->size();
}
pcl::IndicesPtr ind(new std::vector<int>(totalSize));
unsigned int io = 0;
for(unsigned int i=0; i<indices.size(); ++i)
{
for(unsigned int j=0; j<indices[i]->size(); ++j)
{
ind->at(io++) = indices[i]->at(j);
}
}
return ind;
}
pcl::IndicesPtr concatenate(const pcl::IndicesPtr & indicesA, const pcl::IndicesPtr & indicesB)
{
pcl::IndicesPtr ind(new std::vector<int>(*indicesA));
ind->resize(ind->size()+indicesB->size());
unsigned int oi = (unsigned int)indicesA->size();
for(unsigned int i=0; i<indicesB->size(); ++i)
{
ind->at(oi++) = indicesB->at(i);
}
return ind;
}
void savePCDWords(
const std::string & fileName,
const std::multimap<int, pcl::PointXYZ> & words,
const Transform & transform)
{
if(words.size())
{
pcl::PointCloud<pcl::PointXYZ> cloud;
cloud.resize(words.size());
int i=0;
for(std::multimap<int, pcl::PointXYZ>::const_iterator iter=words.begin(); iter!=words.end(); ++iter)
{
cloud[i++] = transformPoint(iter->second, transform);
}
pcl::io::savePCDFile(fileName, cloud);
}
}
void savePCDWords(
const std::string & fileName,
const std::multimap<int, cv::Point3f> & words,
const Transform & transform)
{
if(words.size())
{
pcl::PointCloud<pcl::PointXYZ> cloud;
cloud.resize(words.size());
int i=0;
for(std::multimap<int, cv::Point3f>::const_iterator iter=words.begin(); iter!=words.end(); ++iter)
{
cv::Point3f pt = transformPoint(iter->second, transform);
cloud[i++] = pcl::PointXYZ(pt.x, pt.y, pt.z);
}
pcl::io::savePCDFile(fileName, cloud);
}
}
cv::Mat loadBINScan(const std::string & fileName)
{
cv::Mat output;
long bytes = UFile::length(fileName);
if(bytes)
{
int dim = 4;
UASSERT(bytes % sizeof(float) == 0);
size_t num = bytes/sizeof(float);
UASSERT(num % dim == 0);
output = cv::Mat(1, num/dim, CV_32FC(dim));
// load point cloud
FILE *stream;
stream = fopen (fileName.c_str(),"rb");
size_t actualReadNum = fread(output.data,sizeof(float),num,stream);
UASSERT(num == actualReadNum);
fclose(stream);
}
return output;
}
pcl::PointCloud<pcl::PointXYZ>::Ptr loadBINCloud(const std::string & fileName)
{
return laserScanToPointCloud(loadScan(fileName));
}
pcl::PointCloud<pcl::PointXYZ>::Ptr loadBINCloud(const std::string & fileName, int dim)
{
return loadBINCloud(fileName);
}
LaserScan loadScan(const std::string & path)
{
std::string fileName = UFile::getName(path);
if(UFile::getExtension(fileName).compare("bin") == 0)
{
return LaserScan(loadBINScan(path), 0, 0, LaserScan::kXYZI);
}
else
{
pcl::PCLPointCloud2::Ptr cloud(new pcl::PCLPointCloud2);
if(UFile::getExtension(fileName).compare("pcd") == 0)
{
pcl::io::loadPCDFile(path, *cloud);
}
else // PLY
{
pcl::io::loadPLYFile(path, *cloud);
}
if(cloud->height > 1)
{
cloud->is_dense = false;
}
bool is2D = false;
if(!cloud->data.empty())
{
// If all z values are zeros, we assume it is a 2D scan
int zOffset = -1;
for(unsigned int i=0; i<cloud->fields.size(); ++i)
{
if(cloud->fields[i].name.compare("z") == 0)
{
zOffset = cloud->fields[i].offset;
break;
}
}
if(zOffset>=0)
{
is2D = true;
for (uint32_t row = 0; row < (uint32_t)cloud->height && is2D; ++row)
{
const uint8_t* row_data = &cloud->data[row * cloud->row_step];
for (uint32_t col = 0; col < (uint32_t)cloud->width && is2D; ++col)
{
const uint8_t* msg_data = row_data + col * cloud->point_step;
float z = *(float*)(msg_data + zOffset);
is2D = z == 0.0f;
}
}
}
}
return laserScanFromPointCloud(*cloud, true, is2D);
}
return LaserScan();
}
pcl::PointCloud<pcl::PointXYZ>::Ptr loadCloud(
const std::string & path,
const Transform & transform,
int downsampleStep,
float voxelSize)
{
UASSERT(!transform.isNull());
UDEBUG("Loading cloud (step=%d, voxel=%f m) : %s", downsampleStep, voxelSize, path.c_str());
std::string fileName = UFile::getName(path);
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
if(UFile::getExtension(fileName).compare("bin") == 0)
{
cloud = util3d::loadBINCloud(path); // Assume KITTI velodyne format
}
else if(UFile::getExtension(fileName).compare("pcd") == 0)
{
pcl::io::loadPCDFile(path, *cloud);
}
else
{
pcl::io::loadPLYFile(path, *cloud);
}
int previousSize = (int)cloud->size();
if(downsampleStep > 1 && cloud->size())
{
cloud = util3d::downsample(cloud, downsampleStep);
UDEBUG("Downsampling scan (step=%d): %d -> %d", downsampleStep, previousSize, (int)cloud->size());
}
previousSize = (int)cloud->size();
if(voxelSize > 0.0f && cloud->size())
{
cloud = util3d::voxelize(cloud, voxelSize);
UDEBUG("Voxel filtering scan (voxel=%f m): %d -> %d", voxelSize, previousSize, (int)cloud->size());
}
if(transform.isIdentity())
{
return cloud;
}
return util3d::transformPointCloud(cloud, transform);
}
LaserScan deskew(
const LaserScan & input,
double inputStamp,
const rtabmap::Transform & velocity)
{
if(velocity.isNull())
{
UERROR("velocity should be valid!");
return LaserScan();
}
if(!input.hasTime())
{
UERROR("input scan doesn't have a \"time\" channel! Supported formats: \"%s\", \"%s\".",
LaserScan::formatName(LaserScan::kXYZIT).c_str(),
LaserScan::formatName(LaserScan::kXYZIRT).c_str());
return LaserScan();
}
if(input.empty())
{
UERROR("input scan is empty!");
return LaserScan();
}
int offsetTime = input.getTimeOffset();
// Get latest timestamp
double firstStamp;
double lastStamp;
firstStamp = inputStamp + input.data().ptr<float>(0, 0)[offsetTime];
lastStamp = inputStamp + input.data().ptr<float>(0, input.size()-1)[offsetTime];
if(lastStamp <= firstStamp)
{
UERROR("First and last stamps in the scan are the same!");
return LaserScan();
}
rtabmap::Transform firstPose;
rtabmap::Transform lastPose;
float vx,vy,vz, vroll,vpitch,vyaw;
velocity.getTranslationAndEulerAngles(vx,vy,vz, vroll,vpitch,vyaw);
// 1- The pose of base frame in odom frame at first stamp
// 2- The pose of base frame in odom frame at last stamp
double dt1 = firstStamp - inputStamp;
double dt2 = lastStamp - inputStamp;
firstPose = rtabmap::Transform(vx*dt1, vy*dt1, vz*dt1, vroll*dt1, vpitch*dt1, vyaw*dt1);
lastPose = rtabmap::Transform(vx*dt2, vy*dt2, vz*dt2, vroll*dt2, vpitch*dt2, vyaw*dt2);
if(firstPose.isNull())
{
UERROR("Could not get transform between stamps %f and %f!",
firstStamp,
inputStamp);
return LaserScan();
}
if(lastPose.isNull())
{
UERROR("Could not get transform between stamps %f and %f!",
lastStamp,
inputStamp);
return LaserScan();
}
double stamp;
UTimer processingTime;
double scanTime = lastStamp - firstStamp;
// Preserve ring when input carries it (kXYZIRT): the geometric channel is
// still meaningful after deskewing. Per-point time is zeroed because all
// points share the same pose after correction.
const bool preserveRing = input.hasRing();
const int offsetRing = input.getRingOffset();
const LaserScan::Format outputFormat = preserveRing ? LaserScan::kXYZIRT : LaserScan::kXYZI;
const int outputChannels = preserveRing ? 6 : 4;
cv::Mat output(1, input.size(), CV_32FC(outputChannels));
int offsetIntensity = input.getIntensityOffset();
bool isLocalTransformIdentity = input.localTransform().isIdentity();
Transform localTransformInv = input.localTransform().inverse();
bool timeOnColumns = input.data().cols > input.data().rows;
int oi = 0;
if(timeOnColumns)
{
// t1 t2 ...
// ring1 ring1 ...
// ring2 ring2 ...
// ring3 ring4 ...
// ring4 ring3 ...
for(int u=0; u<input.data().cols; ++u)
{
const float * inputPtr = input.data().ptr<float>(0, u);
stamp = inputStamp + inputPtr[offsetTime];
rtabmap::Transform transform = firstPose.interpolate((stamp-firstStamp) / scanTime, lastPose);
for(int v=0; v<input.data().rows; ++v)
{
inputPtr = input.data().ptr<float>(v, u);
pcl::PointXYZ pt(inputPtr[0],inputPtr[1],inputPtr[2]);
if(pcl::isFinite(pt))
{
if(!isLocalTransformIdentity)
{
pt = rtabmap::util3d::transformPoint(pt, input.localTransform());
}
pt = rtabmap::util3d::transformPoint(pt, transform);
if(!isLocalTransformIdentity)
{
pt = rtabmap::util3d::transformPoint(pt, localTransformInv);
}
float * dataPtr = output.ptr<float>(0, oi++);
dataPtr[0] = pt.x;
dataPtr[1] = pt.y;
dataPtr[2] = pt.z;
dataPtr[3] = inputPtr[offsetIntensity];
if(preserveRing)
{
dataPtr[4] = inputPtr[offsetRing];
dataPtr[5] = 0.0f;
}
}
}
}
}
else // time on rows
{
// t1 ring1 ring2 ring3 ring4
// t2 ring1 ring2 ring3 ring4
// t3 ring1 ring2 ring3 ring4
// t4 ring1 ring2 ring3 ring4
// ... ... ... ... ...
for(int v=0; v<input.data().rows; ++v)
{
const float * inputPtr = input.data().ptr<float>(v, 0);
stamp = inputStamp + inputPtr[offsetTime];
rtabmap::Transform transform = firstPose.interpolate((stamp-firstStamp) / scanTime, lastPose);
for(int u=0; u<input.data().cols; ++u)
{
inputPtr = input.data().ptr<float>(v, u);
pcl::PointXYZ pt(inputPtr[0],inputPtr[1],inputPtr[2]);
if(pcl::isFinite(pt))
{
if(!isLocalTransformIdentity)
{
pt = rtabmap::util3d::transformPoint(pt, input.localTransform());
}
pt = rtabmap::util3d::transformPoint(pt, transform);
if(!isLocalTransformIdentity)
{
pt = rtabmap::util3d::transformPoint(pt, localTransformInv);
}
float * dataPtr = output.ptr<float>(0, oi++);
dataPtr[0] = pt.x;
dataPtr[1] = pt.y;
dataPtr[2] = pt.z;
dataPtr[3] = inputPtr[offsetIntensity];
if(preserveRing)
{
dataPtr[4] = inputPtr[offsetRing];
dataPtr[5] = 0.0f;
}
}
}
}
}
output = cv::Mat(output, cv::Range::all(), cv::Range(0, oi));
UDEBUG("Lidar deskewing time=%fs", processingTime.elapsed());
return LaserScan(output, input.maxPoints(), input.rangeMax(), outputFormat, input.localTransform());
}
}
}