Human face classification using Deep Learning

Author: Tsotne Nadaraia
Keywords: Deep Learning, Classification
Annotation:

Deep Learning is a subfield of machine learning, inspired by the structure of the human brain. Over the last decade, along with technological progress, Deep Learning became most popular and most demanded subfield of machine learning. In this thesis, had been used different methods of Deep Learning for human face classification problem. We have 3 classes for classification: child, woman, man. The main dataset used is UTKFace dataset. The final test results on the dataset generated a classification accuracy of 92%. This result is promising enough and in future we can work on improving our result.