Head Classification based on Convolutional Neural Network

In this paper, we study the head classification using Convolutional Neural Network (CNN). We study the effect of optimizer in various network architecture. We use INRIA datasets and python programming language. We study and evaluate 2 models and 8 optimizers. We evaluate the validation accuracy and training computation time in one epoch. In our experiment result, except optimizers of SGD and Adadelta, the validation accuracy are good. Their performance are above 90%. The average training time of 1 epoch is 3 second.