new_to_linux's profile - activity

yess...It's very difficult to figure out if it's a real person or a photo using a single camera, want to use a 3D / RGBD camera or stereo camera or similar...thanks for your reply..

Hello Everyone,

finally i have complete the small sample project on image based recognition using all source material present on this site this is very helpfull to me for understanding and clearing my many doubt...

but i have one issue. i have trained the images of near 5 person each have 15 images per person in my training text file and for recognition process there is person standing infront of live camera so it will check for recognition....but if i put any one of passport size photo infront camera or show mobile image infront of camera it also should give the correct result.....but this is a proxy how should i recover???as a anyone having a paasport or show the big image of absent person on mobile infront of camera so obiviously it will recognized.....

please suggest any answer...

i have write the below code for reading two images from folder and apply fisherface training algorithm on them there are two person images that i have read....can anyone tell me how to proceed the code for how to recognised person between these two images ???

const char *facerecAlgorithm = "FaceRecognizer.Fisherfaces";

int main() { string facerecAlgorithm = "FaceRecognizer.Fisherfaces"; Ptr<facerecognizer> model;

 // Make sure the "contrib" module is dynamically loaded at runtime.
 // Requires OpenCV v2.4.1 or later (from June 2012), otherwise the FaceRecognizer will not compile or run!
 bool haveContribModule = initModule_contrib();
 if (!haveContribModule) 
 {
    cerr << "ERROR: The 'contrib' module is needed for FaceRecognizer but has not been loaded into OpenCV!" << endl;
    exit(1);
 }
 else
 {
    cout <<  "\ncontrib module is successfully loaded" << std::endl;    
 }


 model = Algorithm::create<FaceRecognizer>(facerecAlgorithm);

 if (model.empty()) 
 {
cerr << "ERROR: The FaceRecognizer [" << facerecAlgorithm;
cerr << "] is not available in your version of OpenCV. ";
cerr << "Please update to OpenCV v2.4.1 or newer." << endl;
exit(1);
 }
 else
 {
    cout <<  "\nFisherFaceRecognizer is successfully loaded" << std::endl;  
 }

 cout <<  "\nFisherFaceRecognizer training starting......" << std::endl;

// holds images and labels
vector<Mat> images;
vector<int> labels;

// images for first person
images.push_back(imread("1.jpg", CV_LOAD_IMAGE_GRAYSCALE)); labels.push_back(0);
images.push_back(imread("3.jpg", CV_LOAD_IMAGE_GRAYSCALE)); labels.push_back(0);
images.push_back(imread("5.jpg", CV_LOAD_IMAGE_GRAYSCALE)); labels.push_back(0);

// images for second person
images.push_back(imread("2.jpg", CV_LOAD_IMAGE_GRAYSCALE)); labels.push_back(1);
images.push_back(imread("4.jpg", CV_LOAD_IMAGE_GRAYSCALE)); labels.push_back(1);

// This is the common interface to train all of the available cv::FaceRecognizer
// implementations
 model->train(various_images, labels);

 cout <<  "\nFisherFaceRecognizer training finished......" << std::endl;
return 0;

}

i have writeen the code for store images in vector array and train it but i want to see the images is stored or not how to check this?

Mat image1 = imread("1.jpg");

Mat image2 = imread("2.jpg");

vector<mat> various_images;

vector<int> labels;

various_images.push_back(image1);

labels.push_back(0);

various_images.push_back(image2);

labels.push_back(1);

model->train(various_images, labels);

if i doing this but not show any image ==>

namedWindow( "Vector of imgs", WINDOW_AUTOSIZE );// Create a window for display.

imshow("Vector of imgs",various_images[0]);

got the output

Algorithms: 28

BackgroundSubtractor.GMG

BackgroundSubtractor.MOG

BackgroundSubtractor.MOG2

CLAHE

DenseOpticalFlow.DualTVL1

DescriptorMatcher.BFMatcher

DescriptorMatcher.FlannBasedMatcher

FaceRecognizer.Eigenfaces

FaceRecognizer.Fisherfaces

FaceRecognizer.LBPH

Feature2D.BRIEF

Feature2D.BRISK

Feature2D.Dense Feature2D.FAST

Feature2D.FASTX

Feature2D.FREAK

Feature2D.GFTT

Feature2D.Grid

Feature2D.HARRIS

Feature2D.MSER

Feature2D.ORB

Feature2D.STAR

Feature2D.SimpleBlob

GeneralizedHough.POSITION

GeneralizedHough.POSITION_ROTATION

GeneralizedHough.POSITION_SCALE

GeneralizedHough.POSITION_SCALE_ROTATION

StatModel.EM

i have doubt on output of program because for my information the available algorithm in OPENCV 2.4.1 is FaceRecognizer.Eigenfaces FaceRecognizer.Fisherfaces FaceRecognizer.LBPH

hello everyone, i want to see available algorithms in my OPENCV version 2.4.1 for that i have written a code below

vector<string> algorithms;
Algorithm::getList(algorithms);
cout << "Algorithms: " << algorithms.size() << endl;
for (int i=0; i<algorithms.size(); i++)
{
  cout << algorithms[i] << endl;
}

and execute this program it give me following output ==>

Algorithms: 5

CLAHE

GeneralizedHough.POSITION

GeneralizedHough.POSITION_ROTATION

GeneralizedHough.POSITION_SCALE

GeneralizedHough.POSITION_SCALE_ROTATION

is this output is correct or is other any way to find out avaliable algorithm in opencv??

i am working on face recognition on images that stored in my current folder i have done with capture,detect and save in current folder but how to recognized two images cant get any solution??

hello, how to train data using given dataset ?? i have dataset of many images.....

how collecting faces for training ??

hello everyone,

i have done with capture,detect and save the current capture image in folder but can anyone tell me how to do recognition for multiple images??

i have read the above website but cant get it in details can anyone have source sample code for comparing two images ...this two images allready in my project working directory..i have the code of image load and detect but i am cunfusing in comparing two images ??

how to compare two faces??

@berak thanks for reply i have used opencv-2.4.10..and follows step in below given website ==> www.raben.com/book/export/html/3

hello everyone, i am new in opencv i have just installed opecv for ubuntu 12.04 desktop only and now i have to start in face recognition process can u tell me from where i want to start??

i have start (cross compiling opencv) by using LTIB to cross compile opencv for ARM but i got one error where i am stuck

error :could not find glib-genmarshal in your path

glib-2.0 package are failed to build

any one know how to slove it?

Hello, i want to cross compile opencv for arm board....please can any one tell me the procedure how to start??