Learning embedded lines of attraction by self organization for pose and expression invariant face recognition

Seow, Ming–Jung; Alex, Ann Theja; Asari, Vijayan K.

doi:10.1117/1.oe.51.10.107201

Cited by 4 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recognition Rate Sparse Representation-based Classification [55] 99.49 Compressive Sensing based [56] 100 Self organization learning embedded lines of attraction [57] 100 Our Proposed Method 99.93 Table V.1: The comparison with the results in [55,56,57] on CMU AMP Facial Expression database one of them needs to consider choosing the proper parameter in order to balance the accuracy and computation cost. We lose some accuracy but the efficiency is higher.…”

Section: Methodsmentioning

confidence: 99%

Adaptive weighted local textural features for illumination, expression, and occlusion invariant face recognition

Asari

2014

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

I still remember the day when I first talked with Dr. Vijayan K. Asari. He told me about the significance and the spirit of academic research. Without his professional guidance, patient explanation, and continuous encouragement, I could hardly finish my thesis. I would like to thank all of the professors who taught me with the related courses during the two years at University of Dayton, and I sincerely appreciate Dr. Eric J. Balster and Dr. Raúl Ordóñez, who spent time commenting on my thesis and attending to my defense. I would like to thank all the members in Vision Lab and my friends who gave me various ideas and suggestions for improving the performance of my research, as well as helping me review my thesis. I really appreciate their help, and I am very touched by their support. With the support and inspiration of my parents, I have the chance to visit and study in United States. They lead me to the engineering world, and I will do my best to achieve the dream. v

show abstract

Section: Methodsmentioning

confidence: 99%

Adaptive weighted local textural features for illumination, expression, and occlusion invariant face recognition

Asari

2014

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

show abstract

“…Given equation 16, we can embed the normalized coefficients to multiply the weights using the following equation. 11) . .…”

Section: Nonlinear Dimensionality Reductionmentioning

confidence: 97%

“…The nonlinear line attractor (NLA) network, formed by Seow et al 9 is the type we will be focusing on, which has applications in skin color association, pattern association, 10 and pose and expression invariant face recognition. 11 Most recurrent neural networks are totally interconnected, meaning that every node in the network is connected to each other. If we glean information about the structures in the brain, we can see that the neurons are not fully interconnected, but connected only to surrounding neurons.…”

Section: Introductionmentioning

confidence: 99%

Gaussian weighted neighborhood connectivity of nonlinear line attractor for learning complex manifolds

2015

Self Cite

View full text Add to dashboard Cite

The human brain has the capability to process high quantities of data quickly for detection and recognition tasks. These tasks are made simpler by the understanding of data, which intentionally removes redundancies found in higher dimensional data and maps the data onto a lower dimensional space. The brain then encodes manifolds created in these spaces, which reveal a specific state of the system. We propose to use a recurrent neural network, the nonlinear line attractor (NLA) network, for the encoding of these manifolds as specific states, which will draw untrained data towards one of the specific states that the NLA network has encoded. We propose a Gaussian-weighted modular architecture for reducing the computational complexity of the conventional NLA network. The proposed architecture uses a neighborhood approach for establishing the interconnectivity of neurons to obtain the manifolds. The modified NLA network has been implemented and tested on the ElectroOptic Synthetic Vehicle Model Database created by the Air Force Research Laboratory (AFRL), which contains a vast array of high resolution imagery with several different lighting conditions and camera views. It is observed that the NLA network has the capability for representing high dimensional data for the recognition of the objects of interest through its new learning strategy. A nonlinear dimensionality reduction scheme based on singular value decomposition has found to be very effective in providing a low dimensional representation of the dataset. Application of the reduced dimensional space on the modified NLA algorithm would provide fast and more accurate recognition performance for real time applications.

show abstract

Weighted Nonlinear Line Attractor for Complex Manifold Learning

Aspiras

Asari

Sakla

2016

Studies in Computational Intelligence

View full text Add to dashboard Cite

Learning embedded lines of attraction by self organization for pose and expression invariant face recognition

Cited by 4 publications

References 17 publications

Adaptive weighted local textural features for illumination, expression, and occlusion invariant face recognition

Adaptive weighted local textural features for illumination, expression, and occlusion invariant face recognition

Gaussian weighted neighborhood connectivity of nonlinear line attractor for learning complex manifolds

Weighted Nonlinear Line Attractor for Complex Manifold Learning

Contact Info

Product

Resources

About