Age Estimation Using Correlation-Refined Features of Convolutional Neural Network

doi:10.6688/JISE.202111_37(6).0014

CNU IR > College of Smart Living > Dept. of Information Management > Periodical Articles > Item 310902800/34341

Please use this identifier to cite or link to this item: https://ir.cnu.edu.tw/handle/310902800/34341

Title:	Age Estimation Using Correlation-Refined Features of Convolutional Neural Network
Authors:	Lee, Jiann-Shu Chang, Chung-Wen Kao, Teng-Wei Wang, Jing-Wein
Contributors:	Natl Univ Tainan, Dept Comp Sci & Informat Engn Chia Nan Univ Pharm & Sci, Dept Management Informat Sci Natl Kaohsiung Univ Sci & Technol, Inst Photon & Commun
Keywords:	age estimation CNN transfer learning deep learning canonical correlation analysis
Date:	2021
Issue Date:	2023-11-11 11:44:41 (UTC+8)
Publisher:	INST INFORMATION SCIENCE
Abstract:	Age estimation remains challenging because of its high dependence on small facial changes (based on individual propagation, increased wrinkles, and even racial or gender factors). In the past decade, some learning models of neural network based on image analysis have been rapidly developed to overcome such limitations. In this study, we developed a novel method, namely correlation-refined convolutional neural network (CR-CNN), based on some deep learning model (AlexNet). Additional to the parameters in model, the CR-CNN model considers a specific learning network, in which the neuron parameters along with refined facial features at various field-of-view levels have determined through canonical correlation analysis (CCA). Such a novel learning strategy, called low-to -middle-level-features retained transfer learning (LMLFR). Through LMLFR, the feature maps in CNN would be reorganized and join as new layer. That means the maps with high CCA values, in which neurons have high coadaptation with respect to feature-map values, are averaged and flattened; and contrarily, the maps with low CCA values are retained for the low-coadaptation neurons. All refined layers are then subjected to principal component analysis to further reduce dimensionality. At the output layer, classification is executed through support vector regressions (SVR) and Marginal Fisher Analysis (MFA) to overcome the non-Gaussian distributions of refined features on different layers. Experiments were conducted using images obtained from the well-known MORPH dataset, and the results indicated that for age estimation, the proposed model outperformed commonly used methods; the error range was approximately -5 to +5, covering approximately 80% of the learning age range. The proposed model constitutes a novel approach to feature refinement and can potentially become the basis of extensive applications.
Relation:	J INF SCI ENG, v.37, n.6, pp.1435-1448
Appears in Collections:	[Dept. of Information Management] Periodical Articles

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	234	View/Open
JISE.202111_37(6).0014.pdf		536Kb	Adobe PDF	140	View/Open

Loading...