立体图像重定向及其质量评价研究

发布时间：2024-05-27 03:10

　　随着立体图像和视频技术的进步以及国际三维成像标准的不断提高,人类正在开启一个3D视觉应用的新时代。图像处理技术的发展和显示设备的多样化迫切需要在各种设备上显示立体内容。为了适应不同的观看需求,需要调整图像的大小,以便在不同的终端设备上实现良好的显示效果。最简单的图像大小调整方法是均匀缩放,但这种方法往往会导致图像中重要对象的失真,进而影响观看舒适性。与传统的二维视觉内容相比,立体视觉引入了额外的深度维度,为用户带来了更多的享受。然而,这一额外的维度也会给立体视觉成像质量带来一些额外的挑战和制约。因此,有必要在尽可能避免视觉畸变、保持深度、获得最佳图像质量感知的前提下实现图像尺寸的调整,这个过程被称为立体图像重定向。然而,由于目前的立体图像重定向技术还不够成熟,立体图像重定向算子对图像质量、深度感知和视觉舒适度的影响带来了不同的扭曲和失真。失真的立体图像会严重影响用户的视觉体验,甚至会造成一些视觉健康问题。因此,重定向立体图像的质量评价在标立体图像重定向的应用和用户体验质量方面发挥着至关重要的作用。探讨如何建立与人眼主观感知相一致的客观立体图像重定向质量评价模型,是立体图像处理领域的研究...

【文章页数】：100 页

【学位级别】：博士

【文章目录】：
摘要
ABSTRACT
Chapter 1 Introduction
    1.1 Motivation
    1.2 Objectives
    1.3 Research Contributions
    1.4 Thesis Structure
Chapter 2 Background and Related Work
    2.1 Stereoscopic Image Retargeting Techniques
        2.1.1 Stereoscopic Image Retargeting Problem
        2.1.2 Stereoscopic Image Retargeting Operators
        2.1.3 Stereoscopic Image Retargeting Distortion Analysis
        2.1.4 Discussion
    2.2 Image Quality Assessment
        2.2.1 Image Quality Overview
        2.2.2 2D Image Quality Assessment Methods
        2.2.3 3D Image Quality Assessment
        2.2.4 Image Retargeting Quality Assessment
    2.3 Summery
Chapter 3 Subjective Stereoscopic Image Retargeting Quality As-sessment
    3.1 Image Acquisition and Selection
    3.2 Test Stimuli
    3.3 Subjective Testing
    3.4 Analysis of Results
        3.4.1 Measuring the effect of retargeting methods on image quality degradation
        3.4.2 Measuring the effect of scene content on image quality degradation
        3.4.3 The correlation between the different quality aspects and overall quality
    3.5 Summery
Chapter 4 Triangulation-based Objective Quality AssessmentMethod for Stereoscopic Image Retargeting
    4.1 Feature Extraction
        4.1.1 Image quality
        4.1.2 Visual comfort
        4.1.3 Depth perception
    4.2 Feature Fusion
    4.3 Experimental Results
        4.3.1 Database
        4.3.2 Performance comparison with other methods
        4.3.3 Performance of each quality component
        4.3.4 Discussion
    4.4 Summery
Chapter 5 Saliency-based Seam Carving Using Adaptive Segmen-tation
    5.1 Overview
        5.1.1 Energy Function
        5.1.2 Saliency Map
    5.2 Proposed Method
        5.2.1 Adaptive Segmentation
        5.2.2 Saliency-based Seam Carving
        5.2.3 Stereoscopic Case
    5.3 Experimental Results
        5.3.1 Image Retargeting Dataset
        5.3.2 Comparison Experiment
        5.3.3 Ablation Experiment
        5.3.4 Discussion
    5.4 Summery
Chapter 6 Conclusion and Future Work
    6.1 Conclusion
    6.2 Future Work
Bibliography
Acknowledgements
Publications



本文编号：3982690