|
This project
explores the development of techniques for facial
feature classification.
- Classifiers
were developed for 11 different features,
including sex, age, race, and expression.
- The performance
of multiple classification techniques
were compared for almost all features.
The key contribution
of this work is its exploration of synthetic
training data.
- Augmenting
the training set with synthetic faces is shown to improve classification
performance significantly.
|
