Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Psychological Impact and Workload of COVID-19 on Healthcare Workers in China During the Early Time of the Pandemic: A Cross-sectional Study.

Disaster medicine and public health preparedness·2022
Same author

Manipulation of Band Alignment in Two-Dimensional Vertical WSe<sub>2</sub>/BA<sub>2</sub>PbI<sub>4</sub> Ruddlesden-Popper Perovskite Heterojunctions via Defect Engineering.

The journal of physical chemistry letters·2022
Same author

Efficient and Stable Perovskite Solar Cells via CsPF<sub>6</sub> Passivation of Perovskite Film Defects.

The journal of physical chemistry letters·2022
Same author

Development of an Indirect ELISA Kit for Rapid Detection of Varicella-Zoster Virus Antibody by Glycoprotein E.

Frontiers in microbiology·2022
Same author

Adhesive Materials Inspired by Barnacle Underwater Adhesion: Biological Principles and Biomimetic Designs.

Frontiers in bioengineering and biotechnology·2022
Same author

Protective effect of cinnamaldehyde on channel catfish infected by drug-resistant Aeromonas hydrophila.

Microbial pathogenesis·2022

相关实验视频

Updated: May 24, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.6K

使用上下文感知照明恢复网络删除肖像影子

Jiangjian Yu, Ling Zhang, Qing Zhang

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
    |March 3, 2025
    PubMed
    概括

    这项研究引入了上下文感知照明恢复网络 (CIRNet),通过使用背景背景来删除肖像影子. 该方法改善了面部和背景之间的照明和,优于现有的技术.

    科学领域:

    • 计算机视觉 计算机视觉
    • 图像处理 图像处理
    • 人工智能的人工智能

    背景情况:

    • 由于面部表面,肖像影子的去除是复杂的.
    • 现有的方法往往忽略了关键的背景照明线索.
    • 背景背景对于照明和在影子清除中至关重要.

    研究的目的:

    • 提出一个新的网络来删除肖像影子,利用背景背景.
    • 为了提高肖像和背景之间的照明一致性.
    • 为了解决现有的影子去除技术的局限性.

    主要方法:

    • 开发了一个具有上下文意识的照明恢复网络 (CIRNet),有三个阶段:CSRNet,ASRNet和全球融合网络.
    • CSRNet可以减轻初始照明差异.
    • ASRNet使用背景和非影子肖像背景来恢复阴影区域.
    • 全球融合网络适应性地将上下文信息合并为最终结果.

    主要成果:

    • 拟议的CIRNet有效地消除了肖像影子,同时保持了照明和.
    • 该方法利用背景照明信息来改善结果.
    • 已证明能够清除高分辨率的阴影和镜像亮光.

    更多相关视频

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    451
    Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters
    07:05

    Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

    Published on: June 18, 2021

    2.3K

    相关实验视频

    Last Updated: May 24, 2025

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

    Published on: November 30, 2022

    2.6K
    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    451
    Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters
    07:05

    Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters

    Published on: June 18, 2021

    2.3K

    结论:

    • 通过结合上下文信息,CIRNet在肖像影子清除方面取得了重大进展.
    • 开发的真实面部影子数据集是同类的第一个,对研究有价值.
    • 与现有方法相比,该方法取得了优越的定性和定量结果.