The computer nose best

Shelina Jilani, Hassan Ugail, Andrew Logan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Downloads (Pure)

Abstract

The nose is the most central feature on the face which is known to exhibit both gender and ethnic differences. It is a robust feature, invariant to expression and known to contain depth information. In this paper we address the topic of binary ethnicity classificiation from images of the nose, using a novel dataset of South Asian, Pakistani images. To the best of our knowledge, we are one of the first to attempt demographic (ethnicity) based identification based solely on information from the nose.A two-category (Pakistani vs Non-Pakistani) task was used in combination with Deep learning (ResNet) based and VGG-based pre-trained models. A series of experiments were conducted using ResNet-50, ResNet-101, ResNet-152, VGG-Face, VGG-16 and VGG-19, for feature extraction and a Linear Support Vector Machine for classification. The experimental results demonstrate ResNet-50 achieves the highest performance accuracy of 94.1%. In comparison, the highest score for the VGG-based models (VGG-16) was 90.8%. These results demonstrate that information from the nose is sufficient for deep learning models to achieve >90% accuracy on judgements of ethnicity.
Original languageEnglish
Title of host publication2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-7281-2741-5
ISBN (Print)978-1-7281-2742-2
DOIs
Publication statusPublished - 6 Feb 2020

Fingerprint

Support vector machines
Feature extraction
Experiments
Deep learning

Keywords

  • Computer Vision
  • Face Perception
  • Ethnicity
  • Nose
  • Face features

Cite this

Jilani, S., Ugail, H., & Logan, A. (2020). The computer nose best. In 2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA IEEE. https://doi.org/10.1109/SKIMA47702.2019.8982474
Jilani, Shelina ; Ugail, Hassan ; Logan, Andrew. / The computer nose best. 2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA. IEEE, 2020.
@inproceedings{6ae4803caa534dcfa3527b85103b71cd,
title = "The computer nose best",
abstract = "The nose is the most central feature on the face which is known to exhibit both gender and ethnic differences. It is a robust feature, invariant to expression and known to contain depth information. In this paper we address the topic of binary ethnicity classificiation from images of the nose, using a novel dataset of South Asian, Pakistani images. To the best of our knowledge, we are one of the first to attempt demographic (ethnicity) based identification based solely on information from the nose.A two-category (Pakistani vs Non-Pakistani) task was used in combination with Deep learning (ResNet) based and VGG-based pre-trained models. A series of experiments were conducted using ResNet-50, ResNet-101, ResNet-152, VGG-Face, VGG-16 and VGG-19, for feature extraction and a Linear Support Vector Machine for classification. The experimental results demonstrate ResNet-50 achieves the highest performance accuracy of 94.1{\%}. In comparison, the highest score for the VGG-based models (VGG-16) was 90.8{\%}. These results demonstrate that information from the nose is sufficient for deep learning models to achieve >90{\%} accuracy on judgements of ethnicity.",
keywords = "Computer Vision, Face Perception, Ethnicity, Nose, Face features",
author = "Shelina Jilani and Hassan Ugail and Andrew Logan",
note = "Acceptance in SAN AAM: no embargo",
year = "2020",
month = "2",
day = "6",
doi = "10.1109/SKIMA47702.2019.8982474",
language = "English",
isbn = "978-1-7281-2742-2",
booktitle = "2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA",
publisher = "IEEE",

}

Jilani, S, Ugail, H & Logan, A 2020, The computer nose best. in 2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA. IEEE. https://doi.org/10.1109/SKIMA47702.2019.8982474

The computer nose best. / Jilani, Shelina; Ugail, Hassan; Logan, Andrew.

2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA. IEEE, 2020.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - The computer nose best

AU - Jilani, Shelina

AU - Ugail, Hassan

AU - Logan, Andrew

N1 - Acceptance in SAN AAM: no embargo

PY - 2020/2/6

Y1 - 2020/2/6

N2 - The nose is the most central feature on the face which is known to exhibit both gender and ethnic differences. It is a robust feature, invariant to expression and known to contain depth information. In this paper we address the topic of binary ethnicity classificiation from images of the nose, using a novel dataset of South Asian, Pakistani images. To the best of our knowledge, we are one of the first to attempt demographic (ethnicity) based identification based solely on information from the nose.A two-category (Pakistani vs Non-Pakistani) task was used in combination with Deep learning (ResNet) based and VGG-based pre-trained models. A series of experiments were conducted using ResNet-50, ResNet-101, ResNet-152, VGG-Face, VGG-16 and VGG-19, for feature extraction and a Linear Support Vector Machine for classification. The experimental results demonstrate ResNet-50 achieves the highest performance accuracy of 94.1%. In comparison, the highest score for the VGG-based models (VGG-16) was 90.8%. These results demonstrate that information from the nose is sufficient for deep learning models to achieve >90% accuracy on judgements of ethnicity.

AB - The nose is the most central feature on the face which is known to exhibit both gender and ethnic differences. It is a robust feature, invariant to expression and known to contain depth information. In this paper we address the topic of binary ethnicity classificiation from images of the nose, using a novel dataset of South Asian, Pakistani images. To the best of our knowledge, we are one of the first to attempt demographic (ethnicity) based identification based solely on information from the nose.A two-category (Pakistani vs Non-Pakistani) task was used in combination with Deep learning (ResNet) based and VGG-based pre-trained models. A series of experiments were conducted using ResNet-50, ResNet-101, ResNet-152, VGG-Face, VGG-16 and VGG-19, for feature extraction and a Linear Support Vector Machine for classification. The experimental results demonstrate ResNet-50 achieves the highest performance accuracy of 94.1%. In comparison, the highest score for the VGG-based models (VGG-16) was 90.8%. These results demonstrate that information from the nose is sufficient for deep learning models to achieve >90% accuracy on judgements of ethnicity.

KW - Computer Vision

KW - Face Perception

KW - Ethnicity

KW - Nose

KW - Face features

U2 - 10.1109/SKIMA47702.2019.8982474

DO - 10.1109/SKIMA47702.2019.8982474

M3 - Conference contribution

SN - 978-1-7281-2742-2

BT - 2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA

PB - IEEE

ER -

Jilani S, Ugail H, Logan A. The computer nose best. In 2019 13th International Conference on Software, Knowledge, Information Management and Applications (SKIMA. IEEE. 2020 https://doi.org/10.1109/SKIMA47702.2019.8982474