Revision history [back]

probably the best way to do so is using the pretrained openface dnn:

    Net net = Dnn.readNetFromTorch("openface.nn4.small2.v1.t7");
    ...
    //Convert Mat to image batch
    Mat inputBlob = dnn::blobFromImage(image, 1./255, new Size(96,96), new Scalar(), true, false);
    net.setInput(inputBlob);
    Mat feature = net.forward();

then compare features using simple L2 norm:

   double dist = Core.norm(f1,f2);

probably the best way to do so is using the pretrained openface dnn:

    Net net = Dnn.readNetFromTorch("openface.nn4.small2.v1.t7");
    ...
    //Convert Mat to image batch
    Mat inputBlob = dnn::blobFromImage(image, 1./255, new Size(96,96), new Scalar(), true, false);
    net.setInput(inputBlob);
    Mat feature = net.forward();

then compare the resulting (128 float) features using simple L2 norm:

   double dist = Core.norm(f1,f2);

probably the best way to do so is using the pretrained openface dnn:

    Net net = Dnn.readNetFromTorch("openface.nn4.small2.v1.t7");
    ...
    //Convert Mat to image batch
    Mat inputBlob blob = dnn::blobFromImage(image, dnn::blobFromImage(img, 1./255, new Size(96,96), new Scalar(), true, false);
    net.setInput(inputBlob);
net.setInput(blob);
    Mat feature = net.forward();

then compare the resulting (128 float) features using simple L2 norm:

   double dist = Core.norm(f1,f2);