Digital Image Processing

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Discriminative Approach for Age Invariant Face Recognition Using CNN and SVM


Affiliations
1 Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
2 Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
     

   Subscribe/Renew Journal


Face Recognition across age has been a challenging and popular task in the field of Face Recognition. Although many researchers are contributing to the area, there’s a significant gap to fill in. Data collections and feature extraction are the most challenging tasks to be tackled in Age Invariant Face Recognition (AIFR). Feature extraction could be achieved using the Convolutional Neural Networks (CNN) and for classification. But due to unavailability of large, paired datasets, opting for a Deep Neural Network architecture for this problem statement would not result in a good accuracy. Therefore, we propose a CNN-SVM architecture where CNN extracts important features from an image, which is then used by SVM for final classification.

Keywords

Age Invariant Face Recognition (AIFR), CNN, SVM, Tensor Flow, Face Detection.
User
Subscription Login to verify subscription
Notifications
Font Size

  • Zhifeng L., et al., “A Discriminative Model for Age Invariant Face Recognition,” IEEE Transactions on Information Forensics and Security, vol. 6, no. 3, pp. 1028-1037, 2011.

  • M. Sajid, et al., “The Role of Facial Asymmetry in Recognizing AgeSeparated Face Images,” Journal of Computers & Electrical Engineering, pp. 1-12, 2016.

  • J. M. Guo, et al., “Human Face Age Estimation with Adaptive Hybrid Features,” International Conference on System Science and Engineering, 2011.

  • Hunter and B. Tiddeman, “Facial Ageing,” Cambridge University Press, 2012.

  • H. Ling, et al., “Face Verification Across Age Progression Using Discriminative Methods,” IEEE Transactions on Information Forensics and Security, vol. 5, no. 1, pp. 82-91, 2010.

  • Sun Y., et al., “Deepid3: Face Recognition with Very Deep Neural Networks,” arXiv preprint arXiv: 1502.00873, 2015.

  • U. Park, et al., “Age-Invariant Face Recognition”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 5, pp. 947-954, 2010.

  • Xiao, et al., “Biview Face Recognition in the Shape- Texture Domain”, Elseview Journal on Pattern Recognition, vol. 46, no. 7, pp. 1906-1919, 2013.

  • Yang H., et al., “Age Invariant Face Recognition Based on Texture Embedded Discriminative Graph Model”, IEEE International Joint Conference on Biometrics, 2014.

  • Guosheng H., et al., “When Face Recognition Meets with Deep Learning: an Evaluation of Convolutional Neural Networks for Face Recognition,” arXiv preprint arXiv:1504.02351, pp. 4321-4329, 2015.

  • The FG-NET Aging Database, https://yanweifu.github.io/FG_NET_data/


Abstract Views: 122

PDF Views: 0




  • Discriminative Approach for Age Invariant Face Recognition Using CNN and SVM

Abstract Views: 122  |  PDF Views: 0

Authors

HB Chaitanya Bharadwaj
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
BS. Avinash
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
MC. Chethan
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
G. Darshan
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India
R. Kanagavalli
Department of Computer Science and Engineering, Global Academy of Technology, Bengaluru-560098, Karnataka, India

Abstract


Face Recognition across age has been a challenging and popular task in the field of Face Recognition. Although many researchers are contributing to the area, there’s a significant gap to fill in. Data collections and feature extraction are the most challenging tasks to be tackled in Age Invariant Face Recognition (AIFR). Feature extraction could be achieved using the Convolutional Neural Networks (CNN) and for classification. But due to unavailability of large, paired datasets, opting for a Deep Neural Network architecture for this problem statement would not result in a good accuracy. Therefore, we propose a CNN-SVM architecture where CNN extracts important features from an image, which is then used by SVM for final classification.

Keywords


Age Invariant Face Recognition (AIFR), CNN, SVM, Tensor Flow, Face Detection.

References