The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation

Gusna Ikhsan; Nanik Suciati

doi:10.1109/ICITISEE57756.2022.10057739

The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation

Gusna Ikhsan
, Nanik Suciati

Faculty of Information and Communication Technology

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

5 Citations (Scopus)

Abstract

The image-to-image translation is part of image generation, which aims to transform the source image into the target image. The benefit of the image-to-image translation method in the forensic field is to recognize the face of the perpetrator of a crime through facial sketches. Pix2pix is a deep learning architecture for image-to-image translation that implements the Conditional Generative Adversarial Network (CGAN). Pix2pix can generate images well, although the resulting image is blurry in many cases. Modification of the Pix2pix architecture through the addition of edge information has been reported to improve the sharpness of the resulting image generation. This study aims to compare the performance of the modified Pix2pix architecture using various edge extraction methods, namely Laplacian, Sobel, and Prewitt. The comparison of the performance of the Similarity Structural Index (SSIM) on the Chinese University of Hong Kong (CUHK) faces student dataset shows that the modification of the Pix2pix architecture with the addition of Prewitt edge information produces the highest average value of 81.4%.

Original language	English
Title of host publication	Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering
Subtitle of host publication	Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	160-165
Number of pages	6
ISBN (Electronic)	9798350399615
DOIs	https://doi.org/10.1109/ICITISEE57756.2022.10057739
Publication status	Published - 2022
Externally published	Yes
Event	6th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2022 - Virtual, Online, Indonesia Duration: 13 Dec 2022 → 14 Dec 2022

Publication series

Name	Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022

Conference

Conference	6th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2022
Country/Territory	Indonesia
City	Virtual, Online
Period	13/12/22 → 14/12/22

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy
SDG 16 Peace, Justice and Strong Institutions

Keywords

CGAN
Pix2pix
SSIM
image
image generation

Access to Document

10.1109/ICITISEE57756.2022.10057739

Cite this

Ikhsan, G., & Suciati, N. (2022). The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation. In Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022 (pp. 160-165). (Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICITISEE57756.2022.10057739

Ikhsan, Gusna ; Suciati, Nanik. / The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation. Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 160-165 (Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022).

@inproceedings{3fc708c5c5474c3c91f59dd82ee3a327,

title = "The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation",

abstract = "The image-to-image translation is part of image generation, which aims to transform the source image into the target image. The benefit of the image-to-image translation method in the forensic field is to recognize the face of the perpetrator of a crime through facial sketches. Pix2pix is a deep learning architecture for image-to-image translation that implements the Conditional Generative Adversarial Network (CGAN). Pix2pix can generate images well, although the resulting image is blurry in many cases. Modification of the Pix2pix architecture through the addition of edge information has been reported to improve the sharpness of the resulting image generation. This study aims to compare the performance of the modified Pix2pix architecture using various edge extraction methods, namely Laplacian, Sobel, and Prewitt. The comparison of the performance of the Similarity Structural Index (SSIM) on the Chinese University of Hong Kong (CUHK) faces student dataset shows that the modification of the Pix2pix architecture with the addition of Prewitt edge information produces the highest average value of 81.4\%.",

keywords = "CGAN, Pix2pix, SSIM, image, image generation",

author = "Gusna Ikhsan and Nanik Suciati",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 6th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2022 ; Conference date: 13-12-2022 Through 14-12-2022",

year = "2022",

doi = "10.1109/ICITISEE57756.2022.10057739",

language = "English",

series = "Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022",

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

pages = "160--165",

booktitle = "Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering",

address = "United States",

}

Ikhsan, G & Suciati, N 2022, The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation. in Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022. Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022, Institute of Electrical and Electronics Engineers Inc., pp. 160-165, 6th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2022, Virtual, Online, Indonesia, 13/12/22. https://doi.org/10.1109/ICITISEE57756.2022.10057739

The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation. / Ikhsan, Gusna; Suciati, Nanik.
Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 160-165 (Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022).

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

TY - GEN

T1 - The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation

AU - Ikhsan, Gusna

AU - Suciati, Nanik

PY - 2022

Y1 - 2022

N2 - The image-to-image translation is part of image generation, which aims to transform the source image into the target image. The benefit of the image-to-image translation method in the forensic field is to recognize the face of the perpetrator of a crime through facial sketches. Pix2pix is a deep learning architecture for image-to-image translation that implements the Conditional Generative Adversarial Network (CGAN). Pix2pix can generate images well, although the resulting image is blurry in many cases. Modification of the Pix2pix architecture through the addition of edge information has been reported to improve the sharpness of the resulting image generation. This study aims to compare the performance of the modified Pix2pix architecture using various edge extraction methods, namely Laplacian, Sobel, and Prewitt. The comparison of the performance of the Similarity Structural Index (SSIM) on the Chinese University of Hong Kong (CUHK) faces student dataset shows that the modification of the Pix2pix architecture with the addition of Prewitt edge information produces the highest average value of 81.4%.

AB - The image-to-image translation is part of image generation, which aims to transform the source image into the target image. The benefit of the image-to-image translation method in the forensic field is to recognize the face of the perpetrator of a crime through facial sketches. Pix2pix is a deep learning architecture for image-to-image translation that implements the Conditional Generative Adversarial Network (CGAN). Pix2pix can generate images well, although the resulting image is blurry in many cases. Modification of the Pix2pix architecture through the addition of edge information has been reported to improve the sharpness of the resulting image generation. This study aims to compare the performance of the modified Pix2pix architecture using various edge extraction methods, namely Laplacian, Sobel, and Prewitt. The comparison of the performance of the Similarity Structural Index (SSIM) on the Chinese University of Hong Kong (CUHK) faces student dataset shows that the modification of the Pix2pix architecture with the addition of Prewitt edge information produces the highest average value of 81.4%.

KW - CGAN

KW - Pix2pix

KW - SSIM

KW - image

KW - image generation

UR - https://www.scopus.com/pages/publications/85150474344

U2 - 10.1109/ICITISEE57756.2022.10057739

DO - 10.1109/ICITISEE57756.2022.10057739

M3 - Conference contribution

AN - SCOPUS:85150474344

T3 - Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022

SP - 160

EP - 165

BT - Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2022

Y2 - 13 December 2022 through 14 December 2022

ER -

Ikhsan G, Suciati N. The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation. In Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 160-165. (Proceeding - 6th International Conference on Information Technology, Information Systems and Electrical Engineering: Applying Data Sciences and Artificial Intelligence Technologies for Environmental Sustainability, ICITISEE 2022). doi: 10.1109/ICITISEE57756.2022.10057739

The Comparative Study of Adding Edge Information to Pix2pix Architecture for Face Image Generation

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this