[Digikam-devel] Google Soc Face Recognition Project

Sukayneh Mutawa sukayneh.mutawa at gmail.com
Sun Mar 28 10:20:59 BST 2010


I am final year computer science student from Saudi Arabia. My field of
interest is image processing and artificial intelligence. I am currently
doing a remote sensing project titled (supervised classification of
multi-spectral images using neural network). I am proficient in
C/C++/Java/Matlab/Prolog.

I would like to develop and enhance digikam's face recognition capability
along with image tagging.

Is there already a code for face detection/face recognition in digikam?
I am quite new to the OpenCV library.   I have been reading OpenCv in the
last two days.

I think I can Use Eigenfaces method to detect faces/ extract features of the
faces and inject features on neural network for classification.

The advantages of this approach is that we utilize the classification power
of neural network ( can be retrained as new images are identified) and
maintaining speed by limiting the data to face's features.

The neural network library  i am going to use Fast Artificial Neural Network
Library (FANN) available at http://leenissen.dk/fann/html/files/fann-h.html.

What do you think?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/digikam-devel/attachments/20100328/599a28ad/attachment.html>


More information about the Digikam-devel mailing list