I am working on emotion recognition using facial expressions. I have to classify static images into one of the following (happy, surprise, sad, angry, etc.,) using characteristics of the face image. i have cropped the image into various parts such as eyes, nose, eyebrows etc. but i don't know what to consider and store as features of the data set, and in which form it is? if it is some numerical values for each of these features to store, how to get those values?
suggest me some good papers or books to refer? or MATLAB functions?
Face recognition is normally done by running a principal component analysis (PCA) on the whole picture.
One of my students has been using the open source tool Fiji (http://fiji.sc/Fiji) to create data sets for Weka, based on image segmentation. This is done via Fiji: preinstalled plugin: Plugins -> Segmentation -> Trainable Weka Segmentation. A set of filters is provided by the plugin.
Our work in briefly described in:
Conference Paper A Windowing based GPU optimized strategy for the induction o...
- Badges
- Science topic
- Similar topics
- Correction