计算机应用研究2023，Vol.40Issue(12)：3810-3815,6.DOI:10.19734/j.issn.1001-3695.2023.04.0171
基于语义先验和双通道特征提取的图像修复
Image restoration based on semantic prior and dual channel feature extraction
摘要
Abstract
Aiming at the problems of local structure disconnection and inaccurate detail restoration in existing image restora-tion algorithms,this paper proposed an inpainting algorithm based on semantic prior and dual channel feature extraction(SPDCE).The algorithm used semantic prior network to learn the semantic information and context knowledge of the missing area,predicted the missing area,and enhanced the local consistency of the generated image.Then,the dual channel feature extraction network fully mined image information,enhanced the perception and utilization of texture details.Next,the context feature adjustment module could capture and encode rich semantic features at multiple scales,thereby generating more realistic image views and finer texture details.After conducting experimental verification on the datasets CelebA-HQ and Places2,the results show that compared with commonly used algorithms,the SPDCE algorithm improved the peak signal ratio(PSNR)and structural similarity(SSIM)by 1.6 dB to 1.73 dBand3.1％to 9.9％,while L,loss decreased by 15.2％to 27.8％.Ex-perimental results show that the repaired image of the proposed algorithm has more reasonable structure and richer details,and the inpainting effect is better.关键词
图像修复/语义先验/双通道特征提取/特征聚合Key words
image inpainting/semantic priori/dual channel feature extraction/feature aggregation分类
信息技术与安全科学引用本文复制引用
杨云,张小璇,杨欣悦..基于语义先验和双通道特征提取的图像修复[J].计算机应用研究,2023,40(12):3810-3815,6.基金项目
国家自然科学基金资助项目(61971272,61601271) （61971272,61601271）
国家重点研发计划资助项目(2019YFC1520204) （2019YFC1520204）
