Skip to main navigation Skip to search Skip to main content

Edge-enhanced GAN with Vanishing Points for Image Inpainting

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

Abstract

Reconstructing the damaged images with perspective views has an extensive range in the field of image inpainting. However, most existing methods generated inadequately realistic restored images. Accomplishing this problem, we propose an edge-enhanced image generation model considering viewpoints. Our method applies edge map information to guide image generation based on the perspective views of an image using vanishing points detection. Texture synthesis will be presented as post-processing to complete the remaining missing regions. Experiment shows that our approach can generate perspective images with convincing details, such as indoor and outdoor facades.

Original languageEnglish
Title of host publicationProceedings - 2022 23rd ACIS International Summer Virtual Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, SNPD-Summer 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages113-118
Number of pages6
ISBN (Electronic)9798350396379
DOIs
Publication statusPublished - 2022
Externally publishedYes
Event23rd ACIS International Summer Virtual Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, SNPD-Summer 2022 - Kyoto City, Japan
Duration: 4 Jul 2022 → 6 Jul 2022

Publication series

NameProceedings - 2022 23rd ACIS International Summer Virtual Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, SNPD-Summer 2022

Conference

Conference23rd ACIS International Summer Virtual Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, SNPD-Summer 2022
Country/TerritoryJapan
CityKyoto City
Period4/07/22 → 6/07/22

Keywords

  • edge map
  • gan
  • image inpainting
  • vanishing points

Fingerprint

Dive into the research topics of 'Edge-enhanced GAN with Vanishing Points for Image Inpainting'. Together they form a unique fingerprint.

Cite this