bool ImageTransformation::findHomography( const Keypoints& source, const Keypoints& result, const Matches& input, Matches& inliers, cv::Mat& homography)
{
if (input.size() < 8)
return false;
std::vector<cv::Point2f> srcPoints, dstPoints;
const int pointsCount = input.size();
for (int i=0; i<pointsCount; i++)
{
srcPoints.push_back(source[input[i].trainIdx].pt);
dstPoints.push_back(result[input[i].queryIdx].pt);
}
std::vector<unsigned char> status;
cv::findHomography(srcPoints, dstPoints, CV_FM_RANSAC, 3, status);
inliers.clear();
for (int i=0; i<pointsCount; i++)
{
if (status[i])
{
inliers.push_back(input[i]);
}
}
return true;
}
bool computeMatchesDistanceStatistics(const Matches& matches, float& meanDistance, float& stdDev)
{
if (matches.empty())
return false;
std::vector<float> distances(matches.size());
for (size_t i=0; i<matches.size(); i++)
distances[i] = matches[i].distance;
cv::Scalar mean, dev;
cv::meanStdDev(distances, mean, dev);
meanDistance = static_cast<float>(mean.val[0]);
stdDev = static_cast<float>(dev.val[0]);
return false;
}
/*
* Find exact matches for the semantic descriptor in the database
*/
bool SemanticDescriptor::findExactMatches(Database & db, Matches & matches)
{
SemanticDescriptor compareID;
map<SemanticDescriptor, string>::iterator iter;
for (iter = db.dDB.begin(); iter != db.dDB.end(); ++iter) {
compareID = iter->first;
if (equals(compareID, db)) {
matches.insert(iter->second);
}
}
return matches.size();
}
bool ImageTransformation::findHomography( const Keypoints& source, const Keypoints& result, const Matches& input, Matches& inliers, cv::Mat& homography)
{
if (input.size() < 4)
return false;
const int pointsCount = input.size();
const float reprojectionThreshold = 2;
//Prepare src and dst points
std::vector<cv::Point2f> srcPoints, dstPoints;
for (int i = 0; i < pointsCount; i++)
{
srcPoints.push_back(source[input[i].trainIdx].pt);
dstPoints.push_back(result[input[i].queryIdx].pt);
}
// Find homography using RANSAC algorithm
std::vector<unsigned char> status;
homography = cv::findHomography(srcPoints, dstPoints, cv::RANSAC, reprojectionThreshold, status);
// Warp dstPoints to srcPoints domain using inverted homography transformation
std::vector<cv::Point2f> srcReprojected;
cv::perspectiveTransform(dstPoints, srcReprojected, homography.inv());
// Pass only matches with low reprojection error (less than reprojectionThreshold value in pixels)
inliers.clear();
for (int i = 0; i < pointsCount; i++)
{
cv::Point2f actual = srcPoints[i];
cv::Point2f expect = srcReprojected[i];
cv::Point2f v = actual - expect;
float distanceSquared = v.dot(v);
if (/*status[i] && */distanceSquared <= reprojectionThreshold * reprojectionThreshold)
{
inliers.push_back(input[i]);
}
}
// Test for bad case
if (inliers.size() < 4)
return false;
// Now use only good points to find refined homography:
std::vector<cv::Point2f> refinedSrc, refinedDst;
for (int i = 0; i < inliers.size(); i++)
{
refinedSrc.push_back(source[inliers[i].trainIdx].pt);
refinedDst.push_back(result[inliers[i].queryIdx].pt);
}
// Use least squares method to find precise homography
cv::Mat homography2 = cv::findHomography(refinedSrc, refinedDst, 0, reprojectionThreshold);
// Reproject again:
cv::perspectiveTransform(dstPoints, srcReprojected, homography2.inv());
inliers.clear();
for (int i = 0; i < pointsCount; i++)
{
cv::Point2f actual = srcPoints[i];
cv::Point2f expect = srcReprojected[i];
cv::Point2f v = actual - expect;
float distanceSquared = v.dot(v);
if (distanceSquared <= reprojectionThreshold * reprojectionThreshold)
{
inliers.push_back(input[i]);
}
}
homography = homography2;
return inliers.size() >= 4;
}
bool performEstimation
(
const FeatureAlgorithm& alg,
const ImageTransformation& transformation,
const cv::Mat& sourceImage,
std::vector<FrameMatchingStatistics>& stat
)
{
Keypoints sourceKp;
Descriptors sourceDesc;
cv::Mat gray;
if (sourceImage.channels() == 3)
cv::cvtColor(sourceImage, gray, CV_BGR2GRAY);
else if (sourceImage.channels() == 4)
cv::cvtColor(sourceImage, gray, CV_BGRA2GRAY);
else if(sourceImage.channels() == 1)
gray = sourceImage;
if (!alg.extractFeatures(gray, sourceKp, sourceDesc))
return false;
std::vector<float> x = transformation.getX();
stat.resize(x.size());
const int count = x.size();
cv::Mat transformedImage;
Keypoints resKpReal;
Descriptors resDesc;
Matches matches;
// To convert ticks to milliseconds
const double toMsMul = 1000. / cv::getTickFrequency();
#pragma omp parallel for private(transformedImage, resKpReal, resDesc, matches)
for (int i = 0; i < count; i++)
{
float arg = x[i];
FrameMatchingStatistics& s = stat[i];
transformation.transform(arg, gray, transformedImage);
int64 start = cv::getTickCount();
alg.extractFeatures(transformedImage, resKpReal, resDesc);
// Initialize required fields
s.isValid = resKpReal.size() > 0;
s.argumentValue = arg;
if (!s.isValid)
continue;
if (alg.knMatchSupported)
{
std::vector<Matches> knMatches;
alg.matchFeatures(sourceDesc, resDesc, 2, knMatches);
ratioTest(knMatches, 0.75, matches);
// Compute percent of false matches that were rejected by ratio test
s.ratioTestFalseLevel = (float)(knMatches.size() - matches.size()) / (float) knMatches.size();
}
else
{
alg.matchFeatures(sourceDesc, resDesc, matches);
}
int64 end = cv::getTickCount();
Matches correctMatches;
cv::Mat homography;
bool homographyFound = ImageTransformation::findHomography(sourceKp, resKpReal, matches, correctMatches, homography);
// Some simple stat:
s.isValid = homographyFound;
s.totalKeypoints = resKpReal.size();
s.consumedTimeMs = (end - start) * toMsMul;
// Compute overall percent of matched keypoints
s.percentOfMatches = (float) matches.size() / (float)(std::min(sourceKp.size(), resKpReal.size()));
s.correctMatchesPercent = (float) correctMatches.size() / (float)matches.size();
// Compute matching statistics
computeMatchesDistanceStatistics(correctMatches, s.meanDistance, s.stdDevDistance);
}
return true;
}
int main() {
cv::VideoCapture cap("../data/guitar_scene.mp4");
if(!cap.isOpened()) {
cerr << "Camera/Video was not opened\n";
return 1;
}
//SystemAlgorithms algorithms(new SIFTDetector(500), new SIFTExtractor, new BruteForceMatcher(cv::NORM_L2), new LucasKanadeAlgorithm);
SystemAlgorithms algorithms = SystemAlgorithms::Create(false, true);
HybridTracker tracker(algorithms);
Marker target = tracker.Registry(PreMarker("../data/guitar_object.jpg", nullptr));
Size2i targetSize = target.GetSize();
vector<cv::Point2f> targetCorner(4);
targetCorner[0] = cv::Point2f(0.0, 0.0);
targetCorner[3] = cv::Point2f(0.0, targetSize.height);
targetCorner[2] = cv::Point2f(targetSize.width, targetSize.height);
targetCorner[1] = cv::Point2f(targetSize.width, 0.0);
uint numFrames = 0;
double time = (double) cv::getTickCount();
Frame frame;
Matches matches;
while(true) {
cap >> frame.image;
if(frame.image.empty() and numFrames == 0) continue;
if(frame.image.empty()) break;
numFrames++;
tracker.Update(frame);
matches = tracker.Find(target, frame);
if(matches.size() > 20) target.SetLost(false);
else target.SetLost(true);
if(matches.size() >= 4) {
vector<Point2f> corners;
Mat homography = cv::findHomography(matches.targetPts(), matches.scenePts(), cv::RANSAC, 4);
cv::perspectiveTransform(targetCorner, corners, homography);
if(!corners.empty()) {
cv::line(frame.image, corners[0], corners[1], cv::Scalar(0, 255, 0), 4);
cv::line(frame.image, corners[1], corners[2], cv::Scalar(0, 255, 0), 4);
cv::line(frame.image, corners[2], corners[3], cv::Scalar(0, 255, 0), 4);
cv::line(frame.image, corners[3], corners[0], cv::Scalar(0, 255, 0), 4);
}
}
for(auto p : matches.scenePts()) {
cv::circle(frame.image, p, 3, cv::Scalar(0, 255, 0), 1);
}
cv::imshow("Tracker Test", frame.image);
char key = cv::waitKey(1);
if(key == 0x1B) // ESC
break;
}
time = (double)(cv::getTickCount() - time) / cv::getTickFrequency();
cout << (numFrames/time) << " fps\n";
return 0;
}
int main( int argc, char **argv )
{
float userTime;
struct rusage startTime, stopTime;
getrusage( RUSAGE_SELF, &startTime );
if ( argc != 4 )
{
std::cout << "Usage: " << argv[0] << " <k> <data file> <query file>" << std::endl << std::endl;
return 1;
}
unsigned k = atoi( argv[1] );
std::string line;
std::cout << "Reading the dataset ........ ";
fflush( stdout );
NumTable db;
std::ifstream dict;
dict.open( argv[2], std::ifstream::in );
while ( getline( dict, line ) )
{
Number h;
std::istringstream reader( line );
reader >> h;
db.push_back( h );
}
NumTable q;
std::ifstream qdict;
qdict.open( argv[3], std::ifstream::in );
while ( getline( qdict, line ) )
{
Number h;
std::istringstream reader( line );
reader >> h;
q.push_back( h );
}
std::cout << "done. " << db.size() << " db hashes and " << q.size() << " query hashes." << std::endl;
std::cout << "Building with " << k << " hamming distance bound ....... ";
fflush( stdout );
HEngine_sn e( k );
e.build( db );
std::cout << "done." << std::endl;
getrusage( RUSAGE_SELF, &stopTime );
userTime =
( (float) ( stopTime.ru_utime.tv_sec - startTime.ru_utime.tv_sec ) ) +
( (float) ( stopTime.ru_utime.tv_usec - startTime.ru_utime.tv_usec ) ) * 1e-6;
std::cout << std::endl;
std::cout << "Building time: " << userTime << " seconds" << std::endl;
std::cout << std::endl;
std::cout << "Searching HEngine matches ......." << std::endl;
getrusage( RUSAGE_SELF, &startTime );
int c = 0;
for ( auto &h: q )
{
Matches res = e.query( h );
c += res.size();
/*for ( auto &d : res )
{
//std::cout << "[" << d.second << "] "<< std::endl;// << HEngine::binStr2Number( h ) << " -> " << HEngine::binStr2Number( d.first ) << std::endl;
c++;
}*/
}
getrusage( RUSAGE_SELF, &stopTime );
userTime =
( (float) ( stopTime.ru_utime.tv_sec - startTime.ru_utime.tv_sec ) ) +
( (float) ( stopTime.ru_utime.tv_usec - startTime.ru_utime.tv_usec ) ) * 1e-6;
std::cout << "found " << c << " total matches. HEngine query time: " << userTime << " seconds" << std::endl << std::endl;
std::cout << "Searching linear matches ......." << std::endl;
getrusage( RUSAGE_SELF, &startTime );
c = 0;
for ( auto &item: db )
{
for ( auto &h: q )
{
unsigned d = HEngine::getHammingDistance( h, item );
if ( d <= k )
{
c++;
}
}
}
getrusage( RUSAGE_SELF, &stopTime );
userTime =
( (float) ( stopTime.ru_utime.tv_sec - startTime.ru_utime.tv_sec ) ) +
( (float) ( stopTime.ru_utime.tv_usec - startTime.ru_utime.tv_usec ) ) * 1e-6;
std::cout << "found " << c << " total matches. Linear query time: " << userTime << " seconds" << std::endl << std::endl;
std::cout << std::endl;
return 0;
}
void Solution::initMyMatches(const Matches& matches){
for (size_t ii = 0; ii < matches.size(); ii++){
myMatches[matches[ii].myDriver.routeId] = matches[ii];
}
}
int main()
{
std::string port=":9000";
int listenQueueBacklog = 400;
if ( FCGX_Init() )
{
exit( 1 );
}
int listen_socket = FCGX_OpenSocket( port.c_str(), listenQueueBacklog );
if ( listen_socket < 0 )
{
exit( 1 );
}
FCGX_Request request;
if ( FCGX_InitRequest( &request, listen_socket, 0 ) )
{
exit( 1 );
}
std::string header = "Content-type: text/html\r\n\r\n";
NumTable db;
std::map<Number,Number> posts;
try
{
sql::Driver *driver;
sql::Connection *con;
sql::Statement *stmt;
sql::ResultSet *res;
driver = get_driver_instance();
con = driver->connect( "tcp://127.0.0.1:3306", "wiki_bot", "31415" );
con->setSchema( "orpheus" );
stmt = con->createStatement();
res = stmt->executeQuery( "SELECT hash, track_id FROM hashes_all" );
while ( res->next() )
{
Number h = res->getUInt64( "hash" );
db.push_back( h );
posts[h] = res->getUInt64( "track_id" );
}
delete res;
delete stmt;
delete con;
}
catch ( sql::SQLException &e )
{
std::cout << "# ERR: SQLException in " << __FILE__;
std::cout << "(" << __FUNCTION__ << ") on line " << __LINE__ << std::endl;
std::cout << "# ERR: " << e.what();
std::cout << " (MySQL error code: " << e.getErrorCode();
std::cout << ", SQLState: " << e.getSQLState() << " )" << std::endl;
}
std::cout << "Start building." << std::endl;
HEngine_sn e( 7 );
e.build( db );
std::cout << "Building done." << std::endl;
while ( FCGX_Accept_r( &request ) == 0 )
{
std::cout << "Have request. " << std::endl;
NumTable q;
Number hash;
std::string query = FCGX_GetParam( "QUERY_STRING", request.envp );
std::stringstream ss( query );
while ( ss >> hash )
{
q.push_back( hash );
if ( ss.peek() == ',' )
{
ss.ignore();
}
}
std::string body = header;
int c = 0;
for ( auto &h: q )
{
Matches res = e.query( h );
c += res.size();
for ( auto &r: res )
{
body += std::to_string( posts[r.first] ) + ":" + std::to_string( r.first ) + ":" + std::to_string( r.second ) + "<br/>";
}
}
FCGX_PutS( body.c_str(), request.out );
FCGX_Finish_r( &request );
}
return 0;
}
