DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Ho-Joon | - |
dc.contributor.author | Lee, JoSeph | - |
dc.contributor.author | Yang, Hyun S. | - |
dc.date.accessioned | 2010-03-18T01:42:35Z | - |
dc.date.available | 2010-03-18T01:42:35Z | - |
dc.date.issued | 2007 | - |
dc.identifier.citation | Lecture Notes on Computer Science, Vol.4492, pp.715-723 | en |
dc.identifier.isbn | 978-3-540-72392-9 | - |
dc.identifier.uri | http://hdl.handle.net/10203/17220 | - |
dc.description.abstract | In this paper, a human action recognition method using a hybrid neu- ral network is presented. The method consists of three stages: preprocessing, feature extraction, and pattern classification. For feature extraction, we propose a modified convolutional neural network (CNN) which has a three-dimensional receptive field. The CNN generates a set of feature maps from the action de- scriptors which are derived from a spatiotemporal volume. A weighted fuzzy min-max (WFMM) neural network is used for the pattern classification stage. We introduce a feature selection technique using the WFMM model to reduce the dimensionality of the feature space. Two kinds of relevance factors between features and pattern classes are defined to analyze the salient features. | en |
dc.description.sponsorship | This research is supported by the ubiquitous computing and network project, the Ministry of Information and Communication 21st century frontier R&D program in Korea. | en |
dc.language.iso | en_US | en |
dc.publisher | Springer Verlag (Germany) | en |
dc.title | Human Action Recognition Using a Modified Convolutional Neural Network | en |
dc.type | Article | en |
dc.identifier.doi | 10.1007/978-3-540-72393-6_85 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.