Ask Your Question
1

Recognition Confidence

asked 2013-11-13 11:49:38 -0500

Chochstr gravatar image

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/

edit retag flag offensive close merge delete

Comments

1

if you set the threshold to 0.0, it will discard all neighbours, and return -1 all the time

berak gravatar imageberak ( 2014-01-13 05:47:05 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
1

answered 2014-01-13 04:34:57 -0500

dervish79 gravatar image

updated 2014-01-13 04:36:13 -0500

It 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

There you find too suggestions to improve the accuracy.

edit flag offensive delete link more
Login/Signup to Answer

Question Tools

Stats

Asked: 2013-11-13 11:49:38 -0500

Seen: 478 times

Last updated: Jan 13 '14