Incremental Face Recognition: Hybrid Approach Using Short-Term Memory and Long-Term Memory

To overcome the plasticity-stability dilemma in incremental face recognition algorithms, we propose a model that employs --- short-term memory (STM) and long-term memory (LTM) based on the Atkinson theory. During the incremental learning the STM can learn the incoming data quickly but due to the limited capacity tends to forget the previously learnt data while trying to learn the new incoming data. Conversely, LTM takes more time to learn the new data but can incorporate the new incoming data effectively while maintaining the previously learnt data. In this paper, we try to improve the learning capability of the STM by using the information present in the LTM by a recall process. To show the effectiveness of the recall process, we evaluated the performance of the STM with and without the recall operation. Experimental results show the successful face recognition performance of the proposed method and the importance of the recall process.

[1]  Miroslaw Bober,et al.  Implementation of Face Recognition Processing Using an Embedded Processor , 2005, J. Robotics Mechatronics.

[2]  Richard C. Atkinson,et al.  Human Memory: A Proposed System and its Control Processes , 1968, Psychology of Learning and Motivation.

[3]  Ilkay Ulusoy,et al.  Generative versus discriminative methods for object recognition , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[4]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[5]  Shaoning Pang,et al.  Incremental learning for online face recognition , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[6]  Anders P. Eriksson,et al.  Is face recognition really a Compressive Sensing problem? , 2011, CVPR 2011.

[7]  Jangsun Baek,et al.  Face recognition using partial least squares components , 2004, Pattern Recognit..

[8]  Haitao Zhao,et al.  A novel incremental principal component analysis and its application for face recognition , 2006, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[9]  Jin Young Choi,et al.  Evolving Logic Networks With Real-Valued Inputs for Fast Incremental Learning , 2009, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[10]  Brian C. Lovell,et al.  Combining Generative and Discriminative Learning for Face Recognition , 2005, Digital Image Computing: Techniques and Applications (DICTA'05).

[11]  Song Ying,et al.  Face recognition based on multi-class SVM , 2009, 2009 Chinese Control and Decision Conference.

[12]  Bernhard Schölkopf,et al.  Estimating the Support of a High-Dimensional Distribution , 2001, Neural Computation.

[13]  Kiyomi Nakamura,et al.  Face recognition evaluation of an association cortex - entorhinal cortex hippocampal formation model by successive learning , 2001, SICE 2001. Proceedings of the 40th SICE Annual Conference. International Session Papers (IEEE Cat. No.01TH8603).

[14]  Allen Y. Yang,et al.  Robust Face Recognition via Sparse Representation , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Stephen Grossberg,et al.  Nonlinear neural networks: Principles, mechanisms, and architectures , 1988, Neural Networks.

[16]  Minho Lee,et al.  Incremental two-dimensional two-directional principal component analysis (I(2D)2PCA) for face recognition , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).