OpenCV Q&A Forum - RSS feedhttp://answers.opencv.org/questions/OpenCV answersenCopyright <a href="http://www.opencv.org">OpenCV foundation</a>, 2012-2018.Mon, 13 Jan 2014 05:47:05 -0600Recognition Confidencehttp://answers.opencv.org/question/23897/recognition-confidence/What is the number mean for the Confidence?
for example: Eigen_model->predict(face_resized, Eigen_Predict, Eigen_Confidence);
and i have the same for LBPH and Fisher algorithms.
I would like to know this to get more accurate predictions because each algorithm is giving me different predictions and none are the correct predictions either.
also here is my project with full code on github.com
https://github.com/Chochstr/OpenCVapp/Wed, 13 Nov 2013 11:49:38 -0600http://answers.opencv.org/question/23897/recognition-confidence/Comment by berak for <p>What is the number mean for the Confidence?
for example: Eigen_model->predict(face_resized, Eigen_Predict, Eigen_Confidence);
and i have the same for LBPH and Fisher algorithms.</p>
<p>I would like to know this to get more accurate predictions because each algorithm is giving me different predictions and none are the correct predictions either. </p>
<p>also here is my project with full code on github.com
<a href="https://github.com/Chochstr/OpenCVapp/">https://github.com/Chochstr/OpenCVapp/</a></p>
http://answers.opencv.org/question/23897/recognition-confidence/?comment=26444#post-id-26444if you [set the threshold to 0.0](https://github.com/Chochstr/OpenCVapp/blob/master/OpenCVapp/OpenCVapp/Recognize.cpp#L219), it will discard *all* neighbours, and return -1 all the timeMon, 13 Jan 2014 05:47:05 -0600http://answers.opencv.org/question/23897/recognition-confidence/?comment=26444#post-id-26444Answer by dervish79 for <p>What is the number mean for the Confidence?
for example: Eigen_model->predict(face_resized, Eigen_Predict, Eigen_Confidence);
and i have the same for LBPH and Fisher algorithms.</p>
<p>I would like to know this to get more accurate predictions because each algorithm is giving me different predictions and none are the correct predictions either. </p>
<p>also here is my project with full code on github.com
<a href="https://github.com/Chochstr/OpenCVapp/">https://github.com/Chochstr/OpenCVapp/</a></p>
http://answers.opencv.org/question/23897/recognition-confidence/?answer=26434#post-id-26434It is the euclidean distance between the test face and the closest face in DB. As I know it is not the optimal measure to get the distance.
This parameter checks how similar the input image is to each training image, and finds the most similar one: the one with the least distance in Euclidean Space. As mentioned in the Servo Magazine article, you might get better results if you use the Mahalanobis space (define USE_MAHALANOBIS_DISTANCE in the code). [Reference](http://www.shervinemami.info/faceRecognition.html)
There you find too suggestions to improve the accuracy.Mon, 13 Jan 2014 04:34:57 -0600http://answers.opencv.org/question/23897/recognition-confidence/?answer=26434#post-id-26434