MATLAB Answers

Training neural network for facial expression classification

4 views (last 30 days)
remo on 28 Feb 2012
Commented: pooja rani on 20 Mar 2019
Hi all,
I have 15 images of happy face,15 image of sad face and 15 image of angry faces.
My problem is i do not know how to train neural network using all this images. I have separate 50 images for testing purposes.
So,how do i train neural network using output from PCA of each image?
PLZ do guide me since my supervisor is pressuring me lately..


Sign in to comment.

Accepted Answer

Greg Heath
Greg Heath on 17 May 2012
Use the vectorization operator, : , to convert each processed image to a column vector. The corresponding target vector is a column of eye(3). Once you have the input and target matrices with 45 columns each, use PATTERNNET
help patternnet
doc patternnet
Hope this helps.


Sign in to comment.

More Answers (1)

vaibhav on 16 May 2012
Are you doing that using PCA??


remo on 18 May 2012
Hi Vaibhav,
i have dropped PCA and adopted LBP for this project
pooja rani
pooja rani on 20 Mar 2019
Hi Remo,
can u share code of training NN classifier with LBP features .

Sign in to comment.


Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!