Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images

Bharindra Kamanditya; Randy Pangestu Kuswara; Muhammad Adi Nugroho; Benyamin Kusumo Putro

doi:10.1109/ICETAS.2018.8629150

Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images

Bharindra Kamanditya, Randy Pangestu Kuswara, Muhammad Adi Nugroho, Benyamin Kusumo Putro

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

From limited two-dimensional recognition, facial recognition has now been developed to be able to recognize three-dimensional face images, which usually involves process of face pose estimation. As the conventional artificial neural networks has shown low recognition rate to this problem, Convolution Neural Network have been the most potential classifier to determine the pose estimation of a three-dimensional face images. Convolution operation is expected to minimize the effect of distortion and disorientation of the object, and able to efficiently reduce the required parameters. Results show that the CNN system could estimate the pose position of the 3D face images with high recognition rate, however, this recognition rate decline significantly for the noisy buried face images, showing the CNN still need improvement to deal with noisy environments.

Original language	English
Title of host publication	2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538679661
DOIs	https://doi.org/10.1109/ICETAS.2018.8629150
Publication status	Published - 28 Jan 2019
Event	5th IEEE International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018 - Bangkok, Thailand Duration: 22 Nov 2018 → 23 Nov 2018

Publication series

Name	2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018

Conference

Conference	5th IEEE International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018
Country	Thailand
City	Bangkok
Period	22/11/18 → 23/11/18

Keywords

convolutional neural network
face recognition
head pose estimation
hyperparameter evaluation
image noises

Access to Document

10.1109/ICETAS.2018.8629150

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images'. Together they form a unique fingerprint.

Cite this

Kamanditya, B., Kuswara, R. P., Nugroho, M. A., & Putro, B. K. (2019). Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images. In 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018 [8629150] (2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICETAS.2018.8629150

Kamanditya, Bharindra ; Kuswara, Randy Pangestu ; Nugroho, Muhammad Adi ; Putro, Benyamin Kusumo. / Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images. 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. (2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018).

@inproceedings{f717a9828a0c4ae4827aab7c2f302dfa,

title = "Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images",

abstract = "From limited two-dimensional recognition, facial recognition has now been developed to be able to recognize three-dimensional face images, which usually involves process of face pose estimation. As the conventional artificial neural networks has shown low recognition rate to this problem, Convolution Neural Network have been the most potential classifier to determine the pose estimation of a three-dimensional face images. Convolution operation is expected to minimize the effect of distortion and disorientation of the object, and able to efficiently reduce the required parameters. Results show that the CNN system could estimate the pose position of the 3D face images with high recognition rate, however, this recognition rate decline significantly for the noisy buried face images, showing the CNN still need improvement to deal with noisy environments.",

keywords = "convolutional neural network, face recognition, head pose estimation, hyperparameter evaluation, image noises",

author = "Bharindra Kamanditya and Kuswara, {Randy Pangestu} and Nugroho, {Muhammad Adi} and Putro, {Benyamin Kusumo}",

year = "2019",

month = jan,

day = "28",

doi = "10.1109/ICETAS.2018.8629150",

language = "English",

series = "2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018",

address = "United States",

note = "5th IEEE International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018 ; Conference date: 22-11-2018 Through 23-11-2018",

}

Kamanditya, B, Kuswara, RP, Nugroho, MA & Putro, BK 2019, Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images. in 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018., 8629150, 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018, Institute of Electrical and Electronics Engineers Inc., 5th IEEE International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018, Bangkok, Thailand, 22/11/18. https://doi.org/10.1109/ICETAS.2018.8629150

Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images. / Kamanditya, Bharindra; Kuswara, Randy Pangestu; Nugroho, Muhammad Adi; Putro, Benyamin Kusumo.

2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. 8629150 (2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images

AU - Kamanditya, Bharindra

AU - Kuswara, Randy Pangestu

AU - Nugroho, Muhammad Adi

AU - Putro, Benyamin Kusumo

PY - 2019/1/28

Y1 - 2019/1/28

N2 - From limited two-dimensional recognition, facial recognition has now been developed to be able to recognize three-dimensional face images, which usually involves process of face pose estimation. As the conventional artificial neural networks has shown low recognition rate to this problem, Convolution Neural Network have been the most potential classifier to determine the pose estimation of a three-dimensional face images. Convolution operation is expected to minimize the effect of distortion and disorientation of the object, and able to efficiently reduce the required parameters. Results show that the CNN system could estimate the pose position of the 3D face images with high recognition rate, however, this recognition rate decline significantly for the noisy buried face images, showing the CNN still need improvement to deal with noisy environments.

AB - From limited two-dimensional recognition, facial recognition has now been developed to be able to recognize three-dimensional face images, which usually involves process of face pose estimation. As the conventional artificial neural networks has shown low recognition rate to this problem, Convolution Neural Network have been the most potential classifier to determine the pose estimation of a three-dimensional face images. Convolution operation is expected to minimize the effect of distortion and disorientation of the object, and able to efficiently reduce the required parameters. Results show that the CNN system could estimate the pose position of the 3D face images with high recognition rate, however, this recognition rate decline significantly for the noisy buried face images, showing the CNN still need improvement to deal with noisy environments.

KW - convolutional neural network

KW - face recognition

KW - head pose estimation

KW - hyperparameter evaluation

KW - image noises

UR - http://www.scopus.com/inward/record.url?scp=85062888878&partnerID=8YFLogxK

U2 - 10.1109/ICETAS.2018.8629150

DO - 10.1109/ICETAS.2018.8629150

M3 - Conference contribution

AN - SCOPUS:85062888878

T3 - 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018

BT - 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th IEEE International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018

Y2 - 22 November 2018 through 23 November 2018

ER -

Kamanditya B, Kuswara RP, Nugroho MA, Putro BK. Convolution Neural Network for Pose Estimation of Noisy Three-Dimensional Face Images. In 2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018. Institute of Electrical and Electronics Engineers Inc. 2019. 8629150. (2018 IEEE 5th International Conference on Engineering Technologies and Applied Sciences, ICETAS 2018). https://doi.org/10.1109/ICETAS.2018.8629150