Restricted Boltzmann Machines for Gender Classification

Abstract

Abstract. This paper deals with automatic feature learning using Gaussian Restricted Boltzmann Machines (GRBM) for the problem of gender recognition in face images. The GRBM is presented together with some practical learning tricks to improve the learning capabilities and speedup the training process. The performance of the features obtained is compared against several linear methods using the same dataset and the same evaluation protocol. The results show a classification accuracy improvement compared with classical linear projection methods. Moreover, in order to increase even more the classification accuracy, we have run some experiments where an SVM is fed with the non-linear mapping obtained by the GRBM in a tandem configuration.

Local Deep Neural Networks for gender recognition

“…These models are able to automatically extract good features from unlabeled data that are useful in supervised tasks like the gender recognition problem (Mansanet et al (2014)). On the other hand, DCNNs models has shown great performance in computer vision tasks by learning from small regions in the visual field, (Krizhevsky et al (2012); Simonyan and Zisserman (2014)) and have been successfully used for face recognition (Taigman et al (2014); Sun et al (2014); Schroff et al (2015)).…”