How to find the top 'x' matches using the facial recognition algorithms

asked 2013-09-04 20:42:59 -0600

InsightfulPeter gravatar image

Hi,

I have an application that successfully finds and identifies faces in images. However, I need to change the application so that instead of it suggesting the most likely face (with associated confidence/distance), it instead provides the top 3 or 5 most likely faces with all their associated confidences. I then perform some other analytics to determine the best match.

I have not found any information in the opencv documentation on how to do this. I'm using an LBPH algorithm and using the "predict" method.

Can anyone provide any guidance on how I can have the algorithm produce the top 'x' matches?

Thanks

edit retag flag offensive close merge delete

Comments

A quick search on this forum has let me to this: http://answers.opencv.org/question/3494/adding-new-method-in-facerecognizer/ I guess this solves the problem. If so i will close this double one down.

StevenPuttemans gravatar imageStevenPuttemans ( 2013-09-05 01:23:23 -0600 )edit
add a comment