基于深度学习的全景图像质量评价研究现状及展望OA
Research Status and Prospects of Omnidirectional Image Quality Assessment Based on Deep Learning
近年来,随着虚拟现实技术的迅速发展,全景图像逐渐受到广泛关注,它在提供沉浸式体验方面具有重要作用.现有的图像客观质量评价方法未能有效评估全景图像的质量,因此为全景图像设计专门的图像客观质量评价方法显得尤为必要.全面总结了基于深度学习的全景图像客观质量评价方法的研究进展.分析了全景图像的特点;根据评价方法的输入图像类型,将全景图像客观质量评价方法划分为四类,分别是等距柱状投影格式的质量评价方法、分段球面投影格式的质量评价方法、立方体投影格式的质量评价方法和视口图像的质量评价方法,并比较了这些方法的原理、特点和性能;总结了全景图像客观质量评价中使用的数据集和评价指标;对全景图像客观质量评价的未来发展方向进行展望,为后续研究提供了切实可行的研究思路.
In recent years,with the rapid development of virtual reality technology,omnidirectional images have gradually gained widespread attention due to their significant role in providing immersive experiences.Existing objective image quality assessment methods have not effectively evaluated the quality of omnidirectional images,making it particularly necessary to design specialized objective quality assessment methods for omnidirectional images.This paper provides a comprehensive summary of the research progress in deep learning-based objective quality assessment methods for omni-directional images.The characteristics of omnidirectional images are analyzed.Based on the input image type of the assess-ment methods,objective quality assessment methods for omnidirectional images are divided into four categories:equirectangular projection format quality assessment methods,segmented spherical projection format quality assessment methods,cubic projection format quality assessment methods,and viewport image quality assessment methods.The prin-ciples,characteristics,and performance of these methods are compared.The datasets and evaluation metrics used in ob-jective omnidirectional image quality assessment are summarized.The future development directions of objective omnidirec-tional image quality assessment are discussed,and practical research ideas for subsequent studies are provided.
田颖哲;董武;陆利坤;马倩;周子镱;张二青
北京印刷学院 信息工程学院,北京 102600北京印刷学院 信息工程学院,北京 102600北京印刷学院 信息工程学院,北京 102600北京印刷学院 信息工程学院,北京 102600北京印刷学院 信息工程学院,北京 102600北京印刷学院 信息工程学院,北京 102600
信息技术与安全科学
全景图像深度学习图像质量评价虚拟现实客观评价
omnidirectional imagedeep learningimage quality assessmentvirtual realityobjective assessment
《计算机科学与探索》 2026 (3)
650-670,21
北京印刷学院校级科研项目(E6202405)北京印刷学院学科建设和研究生教育专项(21090525014,21090325003)北京印刷学院信息与通信工程一级学科博士点培育项目(21090525004)北京印刷学院出版学新兴交叉学科平台建设项目(04190123001/003)北京印刷学院科研平台建设项目(KYCPT202509).This work was supported by the University-Level Scientific Research Project of Beijing Institute of Graphic Communication(E6202405),the Discipline Construction and Postgraduate Education Special Project of Beijing Institute of Graphic Communication(21090525014,21090325003),the First-Level Discipline Doctoral Program Cultivation Project in Information and Communica-tion Engineering of Beijing Institute of Graphic Communication(21090525004),the Emerging Interdisciplinary Platform Construction Project for Publishing Science of Beijing Institute of Graphic Communication(04190123001/003),and the Scientific Research Platform Construction Project of Beijing Institute of Graphic Communication(KYCPT202509).
评论