Diversified Image Inpainting With Transformers and Denoising Iterative Refinement

Image inpainting is a long-standing key problem in the field of computer vision, which aims to fill the missing parts of an image with visually realistic and semantically appropriate content. For a long time, in the research work at home and abroad, how to generate diverse and realistic images is a...

Full description

Saved in:
Bibliographic Details
Main Authors: Shuzhen Xu, Wenlong Xiang, Cuicui Lv, Shuo Wang, Guanhua Liu
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10788702/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850113945598164992
author Shuzhen Xu
Wenlong Xiang
Cuicui Lv
Shuo Wang
Guanhua Liu
author_facet Shuzhen Xu
Wenlong Xiang
Cuicui Lv
Shuo Wang
Guanhua Liu
author_sort Shuzhen Xu
collection DOAJ
description Image inpainting is a long-standing key problem in the field of computer vision, which aims to fill the missing parts of an image with visually realistic and semantically appropriate content. For a long time, in the research work at home and abroad, how to generate diverse and realistic images is a dilemma faced by image inpainting. With the continuous iteration of deep learning technology, transformer feature extraction model and a new generation paradigm, diffusion model, are emerging in vision tasks. Attention-based transformer models can effectively model long-distance dependencies and flexibly design output content. The diffusion model is stable in training, and the quality of its generated images is already better than that of generative adversarial networks. We decompose the inpainting problem into two key steps: diversified pre-generation and high-resolution reconstruction. Firstly, referring to the discretized pixel set, a transformer information association model is designed to sample from the granularity of pixel values, so as to obtain low-resolution results with diverse appearance. Then, a denoising diffusion model is used to reconstruct the high-resolution image, which is a conditional and iterative refinement process. Ultimately, we achieve a set of image restoration methods that produce diverse results and support high-fidelity output.
format Article
id doaj-art-53ceb8e2ef7645e19632db6740004f6c
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-53ceb8e2ef7645e19632db6740004f6c2025-08-20T02:37:02ZengIEEEIEEE Access2169-35362024-01-011218706818708010.1109/ACCESS.2024.351493010788702Diversified Image Inpainting With Transformers and Denoising Iterative RefinementShuzhen Xu0https://orcid.org/0000-0003-0512-8276Wenlong Xiang1Cuicui Lv2https://orcid.org/0000-0002-5673-0729Shuo Wang3https://orcid.org/0009-0001-0768-7920Guanhua Liu4School of Computer and Control Engineering, Yantai University, Yantai, ChinaSchool of Computer and Control Engineering, Yantai University, Yantai, ChinaSchool of Computer and Control Engineering, Yantai University, Yantai, ChinaSchool of Computer and Control Engineering, Yantai University, Yantai, ChinaCollege of Technology and Data, Yantai Nanshan University, Yantai, Longkou, ChinaImage inpainting is a long-standing key problem in the field of computer vision, which aims to fill the missing parts of an image with visually realistic and semantically appropriate content. For a long time, in the research work at home and abroad, how to generate diverse and realistic images is a dilemma faced by image inpainting. With the continuous iteration of deep learning technology, transformer feature extraction model and a new generation paradigm, diffusion model, are emerging in vision tasks. Attention-based transformer models can effectively model long-distance dependencies and flexibly design output content. The diffusion model is stable in training, and the quality of its generated images is already better than that of generative adversarial networks. We decompose the inpainting problem into two key steps: diversified pre-generation and high-resolution reconstruction. Firstly, referring to the discretized pixel set, a transformer information association model is designed to sample from the granularity of pixel values, so as to obtain low-resolution results with diverse appearance. Then, a denoising diffusion model is used to reconstruct the high-resolution image, which is a conditional and iterative refinement process. Ultimately, we achieve a set of image restoration methods that produce diverse results and support high-fidelity output.https://ieeexplore.ieee.org/document/10788702/Image inpaintingdiffusion modeltransformerdiversified inpainting results
spellingShingle Shuzhen Xu
Wenlong Xiang
Cuicui Lv
Shuo Wang
Guanhua Liu
Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
IEEE Access
Image inpainting
diffusion model
transformer
diversified inpainting results
title Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
title_full Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
title_fullStr Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
title_full_unstemmed Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
title_short Diversified Image Inpainting With Transformers and Denoising Iterative Refinement
title_sort diversified image inpainting with transformers and denoising iterative refinement
topic Image inpainting
diffusion model
transformer
diversified inpainting results
url https://ieeexplore.ieee.org/document/10788702/
work_keys_str_mv AT shuzhenxu diversifiedimageinpaintingwithtransformersanddenoisingiterativerefinement
AT wenlongxiang diversifiedimageinpaintingwithtransformersanddenoisingiterativerefinement
AT cuicuilv diversifiedimageinpaintingwithtransformersanddenoisingiterativerefinement
AT shuowang diversifiedimageinpaintingwithtransformersanddenoisingiterativerefinement
AT guanhualiu diversifiedimageinpaintingwithtransformersanddenoisingiterativerefinement