Simultaneous execution of color adjustment and image completion by generative adversarial network

Naofumi Akimoto, Masaki Hayashi, Shuichi Akizuki, Yoshimitsu Aoki

Research output: Contribution to journalArticle

Abstract

In this paper, we address the problem of performing natural paste synthesis by color adjustment and image completion, in order to solve the completion problem that can specify an object appearing in a completion area. We propose a synthesis network that can extract the context features of the input image and reconstruct an image with the feature, making the inserted object appear in the completion region. In addition, we propose a ingenious method to make input images and learning method using Generative Adversarial Network (GAN) that do not require collection of high cost learning data. We show that color adjustment and image completion based on context features are executed at the same time, and natural pasting synthesis can be performed by using these proposal methods.

Original languageEnglish
Pages (from-to)1033-1040
Number of pages8
JournalSeimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering
Volume84
Issue number12
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

Color
Costs

Keywords

  • Color adjustment
  • Computer graphics
  • Computer vision
  • Convolutional neural network
  • Generative adversarial nets
  • Image completion
  • Inpainting
  • Machine learning

ASJC Scopus subject areas

  • Mechanical Engineering

Cite this

Simultaneous execution of color adjustment and image completion by generative adversarial network. / Akimoto, Naofumi; Hayashi, Masaki; Akizuki, Shuichi; Aoki, Yoshimitsu.

In: Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering, Vol. 84, No. 12, 01.01.2018, p. 1033-1040.

Research output: Contribution to journalArticle

@article{1cd99b7dbddd43d69cde7e8924bdec3e,
title = "Simultaneous execution of color adjustment and image completion by generative adversarial network",
abstract = "In this paper, we address the problem of performing natural paste synthesis by color adjustment and image completion, in order to solve the completion problem that can specify an object appearing in a completion area. We propose a synthesis network that can extract the context features of the input image and reconstruct an image with the feature, making the inserted object appear in the completion region. In addition, we propose a ingenious method to make input images and learning method using Generative Adversarial Network (GAN) that do not require collection of high cost learning data. We show that color adjustment and image completion based on context features are executed at the same time, and natural pasting synthesis can be performed by using these proposal methods.",
keywords = "Color adjustment, Computer graphics, Computer vision, Convolutional neural network, Generative adversarial nets, Image completion, Inpainting, Machine learning",
author = "Naofumi Akimoto and Masaki Hayashi and Shuichi Akizuki and Yoshimitsu Aoki",
year = "2018",
month = "1",
day = "1",
doi = "10.2493/jjspe.84.1033",
language = "English",
volume = "84",
pages = "1033--1040",
journal = "Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering",
issn = "0912-0289",
publisher = "Japan Society for Precision Engineering",
number = "12",

}

TY - JOUR

T1 - Simultaneous execution of color adjustment and image completion by generative adversarial network

AU - Akimoto, Naofumi

AU - Hayashi, Masaki

AU - Akizuki, Shuichi

AU - Aoki, Yoshimitsu

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In this paper, we address the problem of performing natural paste synthesis by color adjustment and image completion, in order to solve the completion problem that can specify an object appearing in a completion area. We propose a synthesis network that can extract the context features of the input image and reconstruct an image with the feature, making the inserted object appear in the completion region. In addition, we propose a ingenious method to make input images and learning method using Generative Adversarial Network (GAN) that do not require collection of high cost learning data. We show that color adjustment and image completion based on context features are executed at the same time, and natural pasting synthesis can be performed by using these proposal methods.

AB - In this paper, we address the problem of performing natural paste synthesis by color adjustment and image completion, in order to solve the completion problem that can specify an object appearing in a completion area. We propose a synthesis network that can extract the context features of the input image and reconstruct an image with the feature, making the inserted object appear in the completion region. In addition, we propose a ingenious method to make input images and learning method using Generative Adversarial Network (GAN) that do not require collection of high cost learning data. We show that color adjustment and image completion based on context features are executed at the same time, and natural pasting synthesis can be performed by using these proposal methods.

KW - Color adjustment

KW - Computer graphics

KW - Computer vision

KW - Convolutional neural network

KW - Generative adversarial nets

KW - Image completion

KW - Inpainting

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85057729188&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057729188&partnerID=8YFLogxK

U2 - 10.2493/jjspe.84.1033

DO - 10.2493/jjspe.84.1033

M3 - Article

AN - SCOPUS:85057729188

VL - 84

SP - 1033

EP - 1040

JO - Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering

JF - Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering

SN - 0912-0289

IS - 12

ER -