Visual Servoing Platform  version 3.5.1 under development (2022-07-07)
#include <visp3/core/vpConfig.h>
#include <visp3/core/vpIoTools.h>
#include <visp3/gui/vpDisplayGDI.h>
#include <visp3/gui/vpDisplayOpenCV.h>
#include <visp3/gui/vpDisplayX.h>
#include <visp3/io/vpVideoReader.h>
#include <visp3/mbt/vpMbGenericTracker.h>
#include <visp3/vision/vpKeyPoint.h>
#if (VISP_HAVE_OPENCV_VERSION >= 0x020400)
void learnCube(const vpImage<unsigned char> &I, vpMbGenericTracker &tracker, vpKeyPoint &keypoint_learning, int id)
std::vector<cv::KeyPoint> trainKeyPoints;
double elapsedTime;
keypoint_learning.detect(I, trainKeyPoints, elapsedTime);
std::vector<vpPolygon> polygons;
std::vector<std::vector<vpPoint> > roisPt;
std::pair<std::vector<vpPolygon>, std::vector<std::vector<vpPoint> > > pair = tracker.getPolygonFaces();
polygons = pair.first;
roisPt = pair.second;
std::vector<cv::Point3f> points3f;
vpKeyPoint::compute3DForPointsInPolygons(cMo, cam, trainKeyPoints, polygons, roisPt, points3f);
keypoint_learning.buildReference(I, trainKeyPoints, points3f, true, id);
for (std::vector<cv::KeyPoint>::const_iterator it = trainKeyPoints.begin(); it != trainKeyPoints.end(); ++it) {
vpDisplay::displayCross(I, (int)it->pt.y, (int)it->pt.x, 4, vpColor::red);
int main(int argc, char **argv)
#if (VISP_HAVE_OPENCV_VERSION >= 0x020400)
try {
std::string videoname = "cube.mp4";
for (int i = 0; i < argc; i++) {
if (std::string(argv[i]) == "--name")
videoname = std::string(argv[i + 1]);
else if (std::string(argv[i]) == "--help" || std::string(argv[i]) == "-h") {
std::cout << "\nUsage: " << argv[0] << " [--name <video name>] [--help] [-h]\n" << std::endl;
return 0;
std::string parentname = vpIoTools::getParent(videoname);
std::string objectname = vpIoTools::getNameWE(videoname);
if (!parentname.empty())
objectname = parentname + "/" + objectname;
std::cout << "Video name: " << videoname << std::endl;
std::cout << "Tracker requested config files: " << objectname << ".[init,"
<< "xml,"
<< "cao or wrl]" << std::endl;
std::cout << "Tracker optional config files: " << objectname << ".[ppm]" << std::endl;
bool usexml = false;
if (vpIoTools::checkFilename(objectname + ".xml")) {
tracker.loadConfigFile(objectname + ".xml");
usexml = true;
if (!usexml) {
vpMe me;
cam.initPersProjWithoutDistortion(547, 542, 339, 235);
tracker.setClipping(tracker.getClipping() | vpMbtPolygon::FOV_CLIPPING);
if (vpIoTools::checkFilename(objectname + ".cao"))
tracker.loadModel(objectname + ".cao");
else if (vpIoTools::checkFilename(objectname + ".wrl"))
tracker.loadModel(objectname + ".wrl");
vpKeyPoint keypoint_learning("ORB", "ORB", "BruteForce-Hamming");
keypoint_learning.setDetectorParameter("ORB", "nLevels", 1);
cv::Ptr<cv::ORB> orb_learning = keypoint_learning.getDetector("ORB").dynamicCast<cv::ORB>();
if (orb_learning) {
#if defined(VISP_HAVE_X11)
vpDisplayX display;
#elif defined(VISP_HAVE_GDI)
vpDisplayGDI display;
#elif defined(VISP_HAVE_OPENCV)
vpDisplayOpenCV display;
std::cout << "No image viewer is available..." << std::endl;
return 0;
* Start the part of the code dedicated to object learning from 3 images
std::string imageName[] = {"cube0001.png", "cube0150.png", "cube0200.png"};
vpHomogeneousMatrix initPoseTab[] = {
vpHomogeneousMatrix(0.02143385294, 0.1098083886, 0.5127439561, 2.087159614, 1.141775176, -0.4701291124),
vpHomogeneousMatrix(0.02651282185, -0.03713587374, 0.6873765919, 2.314744454, 0.3492296488, -0.1226054828),
vpHomogeneousMatrix(0.02965448956, -0.07283091786, 0.7253526051, 2.300529617, -0.4286674806, 0.1788761025)};
for (int i = 0; i < 3; i++) {
vpImageIo::read(I, imageName[i]);
if (i == 0) {
display.init(I, 10, 10);
std::stringstream title;
title << "Learning cube on image: " << imageName[i];
vpDisplay::setTitle(I, title.str().c_str());
tracker.setPose(I, initPoseTab[i]);
tracker.display(I, cMo, cam, vpColor::red);
learnCube(I, tracker, keypoint_learning, i);
vpDisplay::displayText(I, 10, 10, "Learning step: keypoints are detected on visible cube faces", vpColor::red);
if (i < 2) {
vpDisplay::displayText(I, 30, 10, "Click to continue the learning...", vpColor::red);
} else {
vpDisplay::displayText(I, 30, 10, "Click to continue with the detection...", vpColor::red);
keypoint_learning.saveLearningData("cube_learning_data.bin", true);
* Start the part of the code dedicated to detection and localization
vpKeyPoint keypoint_detection("ORB", "ORB", "BruteForce-Hamming");
keypoint_detection.setDetectorParameter("ORB", "nLevels", 1);
cv::Ptr<cv::ORB> orb_detector = keypoint_detection.getDetector("ORB").dynamicCast<cv::ORB>();
orb_detector = keypoint_detection.getDetector("ORB").dynamicCast<cv::ORB>();
if (orb_detector) {
keypoint_detection.loadLearningData("cube_learning_data.bin", true);
keypoint_detection.createImageMatching(I, IMatching);
#if defined VISP_HAVE_X11
vpDisplayX display2;
#elif defined VISP_HAVE_GTK
vpDisplayGTK display2;
#elif defined VISP_HAVE_GDI
vpDisplayGDI display2;
vpDisplayOpenCV display2;
display2.init(IMatching, 50, 50, "Display matching between learned and current images");
vpDisplay::setTitle(I, "Cube detection and localization");
double error;
bool click_done = false;
while (!g.end()) {
keypoint_detection.insertImageMatching(I, IMatching);
vpDisplay::displayText(I, 10, 10, "Detection and localization in process...", vpColor::red);
double elapsedTime;
if (keypoint_detection.matchPoint(I, cam, cMo, error, elapsedTime)) {
tracker.setPose(I, cMo);
tracker.display(I, cMo, cam, vpColor::red, 2);
vpDisplay::displayFrame(I, cMo, cam, 0.05, vpColor::none, 3);
keypoint_detection.displayMatching(I, IMatching);
std::vector<vpImagePoint> ransacInliers = keypoint_detection.getRansacInliers();
std::vector<vpImagePoint> ransacOutliers = keypoint_detection.getRansacOutliers();
for (std::vector<vpImagePoint>::const_iterator it = ransacInliers.begin(); it != ransacInliers.end(); ++it) {
vpImagePoint imPt(*it);
imPt.set_u(imPt.get_u() + I.getWidth());
imPt.set_v(imPt.get_v() + I.getHeight());
for (std::vector<vpImagePoint>::const_iterator it = ransacOutliers.begin(); it != ransacOutliers.end(); ++it) {
vpImagePoint imPt(*it);
imPt.set_u(imPt.get_u() + I.getWidth());
imPt.set_v(imPt.get_v() + I.getHeight());
vpDisplay::displayCircle(IMatching, imPt, 4, vpColor::red);
keypoint_detection.displayMatching(I, IMatching);
cam2.initPersProjWithoutDistortion(cam.get_px(), cam.get_py(), cam.get_u0() + I.getWidth(),
cam.get_v0() + I.getHeight());
tracker.setPose(IMatching, cMo);
tracker.display(IMatching, cMo, cam2, vpColor::red, 2);
vpDisplay::displayFrame(IMatching, cMo, cam2, 0.05, vpColor::none, 3);
vpDisplay::displayText(IMatching, 30, 10, "A click to exit.", vpColor::red);
if (vpDisplay::getClick(I, false)) {
click_done = true;
if (vpDisplay::getClick(IMatching, false)) {
click_done = true;
if (!click_done)
} catch (const vpException &e) {
std::cout << "Catch an exception: " << e << std::endl;
std::cout << "Install OpenCV and rebuild ViSP to use this example." << std::endl;
return 0;
Generic class defining intrinsic camera parameters.
void initPersProjWithoutDistortion(double px, double py, double u0, double v0)
static const vpColor red
Definition: vpColor.h:217
static const vpColor none
Definition: vpColor.h:229
static const vpColor green
Definition: vpColor.h:220
Display for windows using GDI (available on any windows 32 platform).
Definition: vpDisplayGDI.h:129
The vpDisplayGTK allows to display image using the GTK 3rd party library. Thus to enable this class G...
Definition: vpDisplayGTK.h:135
The vpDisplayOpenCV allows to display image using the OpenCV library. Thus to enable this class OpenC...
Use the X11 console to display images on unix-like OS. Thus to enable this class X11 should be instal...
Definition: vpDisplayX.h:135
void init(vpImage< unsigned char > &I, int win_x=-1, int win_y=-1, const std::string &win_title="")
static bool getClick(const vpImage< unsigned char > &I, bool blocking=true)
static void display(const vpImage< unsigned char > &I)
static void displayCross(const vpImage< unsigned char > &I, const vpImagePoint &ip, unsigned int size, const vpColor &color, unsigned int thickness=1)
static void setTitle(const vpImage< unsigned char > &I, const std::string &windowtitle)
static void flush(const vpImage< unsigned char > &I)
static void displayFrame(const vpImage< unsigned char > &I, const vpHomogeneousMatrix &cMo, const vpCameraParameters &cam, double size, const vpColor &color=vpColor::none, unsigned int thickness=1, const vpImagePoint &offset=vpImagePoint(0, 0))
static void displayText(const vpImage< unsigned char > &I, const vpImagePoint &ip, const std::string &s, const vpColor &color)
static void displayCircle(const vpImage< unsigned char > &I, const vpImagePoint &center, unsigned int radius, const vpColor &color, bool fill=false, unsigned int thickness=1)
error that can be emited by ViSP classes.
Definition: vpException.h:72
Implementation of an homogeneous matrix and operations on such kind of matrices.
static void read(vpImage< unsigned char > &I, const std::string &filename, int backend=IO_DEFAULT_BACKEND)
Definition: vpImageIo.cpp:148
Class that defines a 2D point in an image. This class is useful for image processing and stores only ...
Definition: vpImagePoint.h:89
unsigned int getWidth() const
Definition: vpImage.h:246
unsigned int getHeight() const
Definition: vpImage.h:188
static bool checkFilename(const std::string &filename)
Definition: vpIoTools.cpp:800
static std::string getNameWE(const std::string &pathname)
Definition: vpIoTools.cpp:1536
static std::string getParent(const std::string &pathname)
Definition: vpIoTools.cpp:1620
Class that allows keypoints detection (and descriptors extraction) and matching thanks to OpenCV libr...
Definition: vpKeyPoint.h:223
void detect(const vpImage< unsigned char > &I, std::vector< cv::KeyPoint > &keyPoints, const vpRect &rectangle=vpRect())
static void compute3DForPointsInPolygons(const vpHomogeneousMatrix &cMo, const vpCameraParameters &cam, std::vector< cv::KeyPoint > &candidates, const std::vector< vpPolygon > &polygons, const std::vector< std::vector< vpPoint > > &roisPt, std::vector< cv::Point3f > &points, cv::Mat *descriptors=NULL)
Definition: vpKeyPoint.cpp:634
unsigned int buildReference(const vpImage< unsigned char > &I)
Definition: vpKeyPoint.cpp:237
static double rad(double deg)
Definition: vpMath.h:117
Real-time 6D object pose tracking using its CAD model.
virtual void getPose(vpHomogeneousMatrix &cMo) const
virtual void getCameraParameters(vpCameraParameters &camera) const
virtual std::pair< std::vector< vpPolygon >, std::vector< std::vector< vpPoint > > > getPolygonFaces(bool orderPolygons=true, bool useVisibility=true, bool clipPolygon=false)
Definition: vpMe.h:61
void setMu1(const double &mu_1)
Definition: vpMe.h:241
void setSampleStep(const double &s)
Definition: vpMe.h:278
void setRange(const unsigned int &r)
Definition: vpMe.h:271
void setMaskSize(const unsigned int &a)
Definition: vpMe.cpp:459
void setNbTotalSample(const int &nb)
Definition: vpMe.h:255
void setMu2(const double &mu_2)
Definition: vpMe.h:248
void setMaskNumber(const unsigned int &a)
Definition: vpMe.cpp:452
void setThreshold(const double &t)
Definition: vpMe.h:300
Class that enables to manipulate easily a video file or a sequence of images. As it inherits from the...
void acquire(vpImage< vpRGBa > &I)
void open(vpImage< vpRGBa > &I)
void setFileName(const std::string &filename)