版权说明 操作指南
首页 > 成果 > 详情

Lightweight structure-guided network with hydra interaction attention and global–local gating mechanism for high-resolution image inpainting

认领
导出
Link by DOI
反馈
分享
QQ微信 微博
成果类型:
期刊论文
作者:
Gui, Yan;Liu, Yaning;Yan, Chen;Kuang, Lidan;Chen, Zhihua
通讯作者:
Gui, Y
作者机构:
[Gui, Yan; Kuang, Lidan; Liu, Yaning; Yan, Chen] Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha 410114, Hunan, Peoples R China.
[Gui, Yan; Kuang, Lidan; Liu, Yaning; Yan, Chen] Changsha Univ Sci & Technol, Hunan Prov Key Lab Intelligent Proc Big Data Trans, Changsha 410114, Hunan, Peoples R China.
[Chen, Zhihua] East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China.
[Gui, Y] Room 1409,Dianyuan Bldg,Chiling Rd 45, Changsha 410076, Hunan, Peoples R China.
通讯机构:
[Gui, Y ] R
Room 1409,Dianyuan Bldg,Chiling Rd 45, Changsha 410076, Hunan, Peoples R China.
语种:
英文
关键词:
High-resolution image inpainting;Structure reconstruction;Hydra interaction attention;Global-local gated transformer
期刊:
Expert Systems with Applications
ISSN:
0957-4174
年:
2025
卷:
272
页码:
126717
基金类别:
National Natural Science Founda-tion of China [62272164, 61972056]; Hunan Provin-cial Natural Science Foundation of China [2023JJ30050]
机构署名:
本校为第一且通讯机构
院系归属:
计算机与通信工程学院
摘要:
Image inpainting has achieved a superior performance boost in inpainting quality with transformers, because of their powerful long-dependency modeling capacity. However, due to quadratically increased computation complexity with spatial resolution, transformers are not suitable for high-resolution image inpainting tasks, especially when attempting to model structure and texture separately. It remains challenging to accurately and reasonably recover the global structures and texture details while maintaining competitive inference efficiency. In ...

反馈

验证码:
看不清楚,换一个
确定
取消

成果认领

标题:
用户 作者 通讯作者
请选择
请选择
确定
取消

提示

该栏目需要登录且有访问权限才可以访问

如果您有访问权限,请直接 登录访问

如果您没有访问权限,请联系管理员申请开通

管理员联系邮箱:yun@hnwdkj.com