Face Recognition using Transferred Deep Learning for Feature Extraction
Abstract
Face recognition systems are a challenging field in computer vision. An important process and key to success is the feature extraction which requires a lot of data and time for learning. Deep learning has proven to be an outstanding method for extracting relevant features of image classification when a huge amount of data is available. However, it is not an easy task for face recognition, which consists of a small number of images per person considered as classes. This paper applies the idea of transferred learning for feature extraction to a face recognition application. The feature extraction part of the trained deep learning model from a different domain is transferred for extracting face features. Then, the multilayer perceptron neural network is used for model evaluation. Experimental results on public face databases show that the proposed method is highly efficient.
Authors: Amornpan Phornchaicharoen, Praisan Padungweang
Published in: Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON) (2019)