Facial Landmark Extraction Scheme Based on Semantic Segmentation

Hyung Joon Kim, Jisoo Park, Hyeon Woo Kim, Een Jun Hwang

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

2 Citations (Scopus)

Abstract

Facial landmark is a set of features that can be distinguished in the human face with the naked eye. Typical facial landmark includes eyes, eyebrows, nose and mouth. It plays an important role in the human-related image analysis. For example, it can be used to determine whether human beings exist in the image, identify who the person is or recognize the orientation of a face when photographing. Methods for detecting facial landmark can be classified into two groups: One group is based on traditional image processing techniques such as Haar-cascade and edge detection. The other group is based on machine learning technique where landmark is detected through training facial features. However, such techniques have shown low accuracy, especially in the exceptional conditions such as low luminance or overlapped face. To overcome this problem, we propose a new facial landmark extraction scheme using deep learning and semantic segmentation and demonstrate that with even small dataset, our scheme can achieve excellent facial landmark extraction performance.

Original languageEnglish
Title of host publication2018 International Conference on Platform Technology and Service, PlatCon 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538647103
DOIs
Publication statusPublished - 2018 Sep 25
Event2018 International Conference on Platform Technology and Service, PlatCon 2018 - Jeju, Korea, Republic of
Duration: 2018 Jan 292018 Jan 31

Other

Other2018 International Conference on Platform Technology and Service, PlatCon 2018
CountryKorea, Republic of
CityJeju
Period18/1/2918/1/31

Fingerprint

Semantics
Edge detection
Image analysis
Learning systems
Luminance
Image processing
Deep learning

Keywords

  • Convolutional Neural Network
  • Facial Landmark
  • Feature Extraction
  • SegNet
  • Semantic Segmentation
  • VGGNet

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Hardware and Architecture
  • Computer Networks and Communications
  • Computer Science Applications

Cite this

Kim, H. J., Park, J., Kim, H. W., & Hwang, E. J. (2018). Facial Landmark Extraction Scheme Based on Semantic Segmentation. In 2018 International Conference on Platform Technology and Service, PlatCon 2018 [8472730] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PlatCon.2018.8472730

Facial Landmark Extraction Scheme Based on Semantic Segmentation. / Kim, Hyung Joon; Park, Jisoo; Kim, Hyeon Woo; Hwang, Een Jun.

2018 International Conference on Platform Technology and Service, PlatCon 2018. Institute of Electrical and Electronics Engineers Inc., 2018. 8472730.

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

Kim, HJ, Park, J, Kim, HW & Hwang, EJ 2018, Facial Landmark Extraction Scheme Based on Semantic Segmentation. in 2018 International Conference on Platform Technology and Service, PlatCon 2018., 8472730, Institute of Electrical and Electronics Engineers Inc., 2018 International Conference on Platform Technology and Service, PlatCon 2018, Jeju, Korea, Republic of, 18/1/29. https://doi.org/10.1109/PlatCon.2018.8472730
Kim HJ, Park J, Kim HW, Hwang EJ. Facial Landmark Extraction Scheme Based on Semantic Segmentation. In 2018 International Conference on Platform Technology and Service, PlatCon 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8472730 https://doi.org/10.1109/PlatCon.2018.8472730
Kim, Hyung Joon ; Park, Jisoo ; Kim, Hyeon Woo ; Hwang, Een Jun. / Facial Landmark Extraction Scheme Based on Semantic Segmentation. 2018 International Conference on Platform Technology and Service, PlatCon 2018. Institute of Electrical and Electronics Engineers Inc., 2018.
@inproceedings{bd0a7737363340b2a49819875edbd9a8,
title = "Facial Landmark Extraction Scheme Based on Semantic Segmentation",
abstract = "Facial landmark is a set of features that can be distinguished in the human face with the naked eye. Typical facial landmark includes eyes, eyebrows, nose and mouth. It plays an important role in the human-related image analysis. For example, it can be used to determine whether human beings exist in the image, identify who the person is or recognize the orientation of a face when photographing. Methods for detecting facial landmark can be classified into two groups: One group is based on traditional image processing techniques such as Haar-cascade and edge detection. The other group is based on machine learning technique where landmark is detected through training facial features. However, such techniques have shown low accuracy, especially in the exceptional conditions such as low luminance or overlapped face. To overcome this problem, we propose a new facial landmark extraction scheme using deep learning and semantic segmentation and demonstrate that with even small dataset, our scheme can achieve excellent facial landmark extraction performance.",
keywords = "Convolutional Neural Network, Facial Landmark, Feature Extraction, SegNet, Semantic Segmentation, VGGNet",
author = "Kim, {Hyung Joon} and Jisoo Park and Kim, {Hyeon Woo} and Hwang, {Een Jun}",
year = "2018",
month = "9",
day = "25",
doi = "10.1109/PlatCon.2018.8472730",
language = "English",
booktitle = "2018 International Conference on Platform Technology and Service, PlatCon 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Facial Landmark Extraction Scheme Based on Semantic Segmentation

AU - Kim, Hyung Joon

AU - Park, Jisoo

AU - Kim, Hyeon Woo

AU - Hwang, Een Jun

PY - 2018/9/25

Y1 - 2018/9/25

N2 - Facial landmark is a set of features that can be distinguished in the human face with the naked eye. Typical facial landmark includes eyes, eyebrows, nose and mouth. It plays an important role in the human-related image analysis. For example, it can be used to determine whether human beings exist in the image, identify who the person is or recognize the orientation of a face when photographing. Methods for detecting facial landmark can be classified into two groups: One group is based on traditional image processing techniques such as Haar-cascade and edge detection. The other group is based on machine learning technique where landmark is detected through training facial features. However, such techniques have shown low accuracy, especially in the exceptional conditions such as low luminance or overlapped face. To overcome this problem, we propose a new facial landmark extraction scheme using deep learning and semantic segmentation and demonstrate that with even small dataset, our scheme can achieve excellent facial landmark extraction performance.

AB - Facial landmark is a set of features that can be distinguished in the human face with the naked eye. Typical facial landmark includes eyes, eyebrows, nose and mouth. It plays an important role in the human-related image analysis. For example, it can be used to determine whether human beings exist in the image, identify who the person is or recognize the orientation of a face when photographing. Methods for detecting facial landmark can be classified into two groups: One group is based on traditional image processing techniques such as Haar-cascade and edge detection. The other group is based on machine learning technique where landmark is detected through training facial features. However, such techniques have shown low accuracy, especially in the exceptional conditions such as low luminance or overlapped face. To overcome this problem, we propose a new facial landmark extraction scheme using deep learning and semantic segmentation and demonstrate that with even small dataset, our scheme can achieve excellent facial landmark extraction performance.

KW - Convolutional Neural Network

KW - Facial Landmark

KW - Feature Extraction

KW - SegNet

KW - Semantic Segmentation

KW - VGGNet

UR - http://www.scopus.com/inward/record.url?scp=85052657183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052657183&partnerID=8YFLogxK

U2 - 10.1109/PlatCon.2018.8472730

DO - 10.1109/PlatCon.2018.8472730

M3 - Conference contribution

AN - SCOPUS:85052657183

BT - 2018 International Conference on Platform Technology and Service, PlatCon 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -