Adaptive Structural Deep Learning to Recognize Kinship Using Families in Wild Multimedia
Autor: | Shin Kamada, Takumi Ichimura |
---|---|
Rok vydání: | 2021 |
Předmět: | |
Zdroj: | Intelligent Decision Technologies ISBN: 9789811627644 KES-IDT |
DOI: | 10.1007/978-981-16-2765-1_46 |
Popis: | Deep learning has a hierarchical network architecture to represent the complicated feature of input patterns. We have developed the adaptive structure learning method of deep belief network (adaptive DBN) that can discover an optimal number of hidden neurons for given input data in a restricted Boltzmann machine (RBM) by neuron generation–annihilation algorithm and can obtain appropriate number of hidden layers in DBN. In this paper, our model is applied to Families in Wild Multimedia (FIW): A multi-modal database for recognizing kinship. The kinship verification is a problem whether two facial images have the blood relatives or not. In this paper, the two facial images are composed into one image to recognize kinship. The classification accuracy for the developed system became higher than the traditional method. |
Databáze: | OpenAIRE |
Externí odkaz: |