Abstract
This paper offers an artificial neural network that recognizes and segments a face and its components (e.g., eyes and mouth) from a complex background. The selective attention model (Fukushima, 1987) has been extended to have two channels of different resolutions. The high-resolution channel can analyze input patterns in detail, but usually lacks the ability to get global information because of small receptive fields of the cells in it. On the other hands, the low-resolution channel, whose cells have large receptive fields, can capture global information but only roughly. The proposed network analyses object by the interaction of both channels. Computer simulation has demonstrated that the network, which has learned only a small number of facial front views, can recognize and segment faces, eyes and mouths correctly from images containing a variety of faces against complex background.
Preview
Unable to display preview. Download preview PDF.
References
Fukushima, K.: Neural network model for selective attention in visual pattern recognition and associative recall. Applied Optics 26[23] (1987) 4985–4992
Fukushima, K., Hashimoto, H.: Recognition and segmentation of components of a face with selective attention. Trans. MICE D-II, J80-D-II[8] (1997) in press
Fukushima, K.: Neocognitron: a hierarchical neural network capable of visual pattern recognition. Neural Networks 1[2] (1988) 119–130
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Fukushima, K., Hashimoto, H. (1997). Recognition and segmentation of components of a face by a multi-resolution neural network. In: Gerstner, W., Germond, A., Hasler, M., Nicoud, JD. (eds) Artificial Neural Networks — ICANN'97. ICANN 1997. Lecture Notes in Computer Science, vol 1327. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020272
Download citation
DOI: https://doi.org/10.1007/BFb0020272
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63631-1
Online ISBN: 978-3-540-69620-9
eBook Packages: Springer Book Archive