Cover Image

The feature parallelism model of visual recognition

In this work, the Feature Parallelism Model of visual recognition, which addresses the parallel nature of the human brain compared to the hierarchal (serial) brain model, was studied. First, its accuracy rate and training time were compared to those of DeepFace, a leading industry algorithm for f...

Full description

Saved in:
Bibliographic Details
Main Authors: Hassan, Marwa Yousif, Shuriye, Abdi Omar, Hassan Abdalla Hashim, Aisha, Salami, Momoh Jimoh Eyiomika, Khalifa, Othman Omran
Format: Article
Language:English
Published: Science & Engineering Research Support Society (SERSC) 2017
Subjects:
T Technology (General)
Online Access:http://irep.iium.edu.my/56655/1/The%20Feature%20Parallelism%20Model%20of%20Visual%20Recognition.pdf
http://irep.iium.edu.my/56655/
http://www.sersc.org/journals/IJMUE/vol12_no2_2017/13.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.iium.irep.56655
record_format dspace
spelling my.iium.irep.566552018-01-31T17:08:36Z http://irep.iium.edu.my/56655/ The feature parallelism model of visual recognition Hassan, Marwa Yousif Shuriye, Abdi Omar Hassan Abdalla Hashim, Aisha Salami, Momoh Jimoh Eyiomika Khalifa, Othman Omran T Technology (General) In this work, the Feature Parallelism Model of visual recognition, which addresses the parallel nature of the human brain compared to the hierarchal (serial) brain model, was studied. First, its accuracy rate and training time were compared to those of DeepFace, a leading industry algorithm for face recognition. Both models were trained using ImageNet object recognition dataset. Accuracy rates were almost the same, around 57% top-1 error rate and 33% top-5 error rate. Training time for feature parallelism model has dropped to 21% less than that of Deep Face. Second, we have investigated feature parallelism model under depth, i.e., when adding more layers along the horizontal axis. We have tested the model with 5, 6, 7, and 8 layers respectively; we found that the best results both in terms of accuracy rates and training time were obtained with the sixlayered model. Although the training time enhancement was only a few milliseconds when going from 5 to 6 layers, it has worsened significantly when going from 6 to 7 layers. In fact the training time has tripled, i.e., training time of the 7- layers model is three times of that of the 6- layers model. It continues to worsen by a fewer rate with the 8- layers model. Similarly, accuracy rate was better with the 6- layers model by about 1% of that of the 5- layers model; however, it has worsened by more than 5% whenever we add more layers above six. We consider those results are biologically plausible, as they conform to the biological fact that the cerebral cortex is organized in 6- layers. We’ve concluded that the organization of parallel processing units into 6- layers, either in our brains or in artificial vision systems, may enhance both processing time and accuracy rates. Science & Engineering Research Support Society (SERSC) 2017 Article REM application/pdf en http://irep.iium.edu.my/56655/1/The%20Feature%20Parallelism%20Model%20of%20Visual%20Recognition.pdf Hassan, Marwa Yousif and Shuriye, Abdi Omar and Hassan Abdalla Hashim, Aisha and Salami, Momoh Jimoh Eyiomika and Khalifa, Othman Omran (2017) The feature parallelism model of visual recognition. International Journal of Multimedia and Ubiquitous Engineering, 12 (2). pp. 171-186. ISSN 1975-0080 http://www.sersc.org/journals/IJMUE/vol12_no2_2017/13.pdf 10.14257/ijmue.2017.12.2.13
institution Universiti Islam Antarabangsa Malaysia
building IIUM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider International Islamic University Malaysia
content_source IIUM Repository (IREP)
url_provider http://irep.iium.edu.my/
language English
topic T Technology (General)
spellingShingle T Technology (General)
Hassan, Marwa Yousif
Shuriye, Abdi Omar
Hassan Abdalla Hashim, Aisha
Salami, Momoh Jimoh Eyiomika
Khalifa, Othman Omran
The feature parallelism model of visual recognition
description In this work, the Feature Parallelism Model of visual recognition, which addresses the parallel nature of the human brain compared to the hierarchal (serial) brain model, was studied. First, its accuracy rate and training time were compared to those of DeepFace, a leading industry algorithm for face recognition. Both models were trained using ImageNet object recognition dataset. Accuracy rates were almost the same, around 57% top-1 error rate and 33% top-5 error rate. Training time for feature parallelism model has dropped to 21% less than that of Deep Face. Second, we have investigated feature parallelism model under depth, i.e., when adding more layers along the horizontal axis. We have tested the model with 5, 6, 7, and 8 layers respectively; we found that the best results both in terms of accuracy rates and training time were obtained with the sixlayered model. Although the training time enhancement was only a few milliseconds when going from 5 to 6 layers, it has worsened significantly when going from 6 to 7 layers. In fact the training time has tripled, i.e., training time of the 7- layers model is three times of that of the 6- layers model. It continues to worsen by a fewer rate with the 8- layers model. Similarly, accuracy rate was better with the 6- layers model by about 1% of that of the 5- layers model; however, it has worsened by more than 5% whenever we add more layers above six. We consider those results are biologically plausible, as they conform to the biological fact that the cerebral cortex is organized in 6- layers. We’ve concluded that the organization of parallel processing units into 6- layers, either in our brains or in artificial vision systems, may enhance both processing time and accuracy rates.
format Article
author Hassan, Marwa Yousif
Shuriye, Abdi Omar
Hassan Abdalla Hashim, Aisha
Salami, Momoh Jimoh Eyiomika
Khalifa, Othman Omran
author_facet Hassan, Marwa Yousif
Shuriye, Abdi Omar
Hassan Abdalla Hashim, Aisha
Salami, Momoh Jimoh Eyiomika
Khalifa, Othman Omran
author_sort Hassan, Marwa Yousif
title The feature parallelism model of visual recognition
title_short The feature parallelism model of visual recognition
title_full The feature parallelism model of visual recognition
title_fullStr The feature parallelism model of visual recognition
title_full_unstemmed The feature parallelism model of visual recognition
title_sort feature parallelism model of visual recognition
publisher Science & Engineering Research Support Society (SERSC)
publishDate 2017
url http://irep.iium.edu.my/56655/1/The%20Feature%20Parallelism%20Model%20of%20Visual%20Recognition.pdf
http://irep.iium.edu.my/56655/
http://www.sersc.org/journals/IJMUE/vol12_no2_2017/13.pdf
_version_ 1643614968152588288
score 13.211869

Similar Items