QoE Modeling and Applications for Multimedia Systems

Release Date:2013-03-27 Author:Wenjun Zeng and Weisi Lin Click:

Improving the quality and experience perceived by the user is fundamental when developing multimedia technologies, products, and services. Quality of experience (QoE) involves subjective perception, user behavior and needs, appropriateness, context, and usability of delivered content. Modeling QoE is critical for enhancing QoE in various multimedia applications. In this special issue, we present the latest developments, trends, challenges, and practices in QoE modeling and applications for multimedia systems. The seven expert papers in this special issue come from academia and industry. They present some of latest developments in QoE modeling and assessment for emerging scenarios such as 3D video, streaming and cloud systems, user generated content, and mobile user experience.


  We start with two papers that address the problems in 3D QoE assessment and perceptually-driven compression. In “Methodologies for Assessing 3D QoE: Standards and Explorative Studies,” Chen et al. describe the fundamentals of existing subjective video quality assessment methods that are the starting point for 3DTV QoE assessment. The authors discuss potential methods for assessing QoE in stereoscopic 3DTV, focusing mainly on multidimensional QoE indicators and common features of subjective assessment. In “3D Perception Algorithms: Towards Perceptually Driven Compression of 3D Video,” Hu et al. highlight the differences in perceptual effects between 2D and 3D video. They then share their ideas about 3D video coding and transmission, taking into consideration 3D visual attention, 3D just-noticeable-difference, and 3D texture synthesis modeling. We hope that these two papers prompt further thinking about emerging 3D signal processing.


  QoE estimation and modeling has been an important tool for improving user experience in multimedia communication systems. In “Estimating Reduced-Reference Video Quality for Quality-Based Streaming Video,” Atzori et al. analyze reduced-reference algorithms for modeling signal distortion, modeling the human visual system, and analyzing the video signal source. The authors then discuss the practical use of these reduced-reference techniques for monitoring and controlling quality in streaming video systems. As the mobile cloud computing paradigm emerges, QoE has become a much more important issue to investigate. In “Human-Centric Composite-Quality Modeling and Assessment for Virtual Desktop Clouds,” Xu et al. propose a novel reference architecture and discuss its use in modeling and assessing objective user QoE within virtual desktop clouds. This architecture avoids the need for expensive and time-consuming subjective evaluation.


  With the widespread use of smartphones, digital cameras, imaging software, photo-sharing sites, and social networks, the amount of user-generated content has grown tremendously. In “Assessing the Quality of User-Generated Content,” Winkler compares the traditional approaches to assessing quality of user-generated content with new approaches. Some sample applications are also discussed. In “An Improved Color Cast Detection Method Based on Ab-Chromaticity Histogram,” Lu et al. propose a new method for evaluating the quality of an image in order to improve color cast detection. This is a necessary step before further image processing, such as white balance, is applied.


  Energy consumption is a big issue for mobile devices and services. In “Battery Voltage Discharge Rate Prediction and Video Content Adaptation in Mobile Devices on 3G Access Networks,” Mkwawa and Sun propose a way of performing visual content adaptation that saves energy. A regression model is used to predict the battery voltage discharge rate in VoIP applications. This is an interesting attempt. Optimizing user experience with a limited battery is challenging for practical system design (starting from algorithm development).
The guest editorial team would like to thank all authors for submitting their high-quality work to this special issue. We would also like to thank the reviewers whose hard work and expert contributions have ensured the quality of this issue. We hope you enjoy reading these fine quality papers.