Advances in statistical methods-based visual quality assessment

  in Special Issue   Posted on November 3, 2017

Information for the Special Issue

Special Issue Call for Papers:

Visual information, represented by various types of images and videos, is omnipresent, substantial, indispensable, diverse and complicated in our daily life. Regardless of being raw or processed, visual information is ultimately received and interpreted by our human beings. To assess the quality of images and videos, some traditional measures like the Peak Signal to Noise Ratio (PSNR) has been widely used. However, the inconsistency between these traditional measures and the human vision system (HVS) has hindered the development of visual information processing. Being aware of this problem, a large number of practitioners from the computer vision and image processing communities have focused on developing new metrics of visual quality assessment (VQA), which are designed perceptually more consistent to the HVS. In early research, they focused on imitating the HVS with the help of psychophysics. Then the trend in research became to treat the HVS as a black box and just imitate its functions. More recently, the practitioners start to exploit the links between statistics and the HVS, which were shaped and developed throughout the evolution of the HVS. In fact, the use of statistics, including the local and global summary statistics, statistical models and statistical machine learning techniques, becomes more and more popular in each constituent module of VQA, no matter there is reference information or not for assessment.

In this context, this special issue aims to call for the state-of-the-art research in the technology, methodology, theory and application of VQA, especially the statistics-related aspects involved in VQA. It also aims to demonstrate the recent efforts made by the relevant researchers in the fields of computer vision, image processing, statistics and machine learning.

We welcome all the relevant, original work, including but not limited to:

  • Statistics for natural scenes.
  • Statistics for specific types of image, such as screen content images.
  • Statistics for specific distortions of image.
  • Spatial and temporal statistics for videos.
  • Statistics-based perceptual features for VQA.
  • Statistical machine learning for VQA.
  • Deep learning for VQA.
  • Hybrid statistical and non-statistical learning for VQA.
  • Statistics-based pooling strategies in VQA.
  • Statistical evaluation of VQA methods.
  • Statistical analysis and interpretation of existing VQA methods.
  • VQA algorithms for image/video compression, denoising, restoration, enhancement, super-resolution, etc.
  • VQA applications in biometrics, medical imaging, remote sensing, security, etc.
  • VQA databases.

 

Important dates

CFP release: 1 November 2017

Manuscript submission: 31 May 2018

First notification: 31 August 2018

Revision: 31 October 2018

Final decision: 30 November 2018

Tentative publication date: February 2019

 

Guest editors

  • Dr Fei Zhou, Tsinghua University, China

Email: flyingzhou@sz.tsinghua.edu.cnflying.zhou@163.com

  • Prof Wenming Yang, Tsinghua University, China

Email: yang.wenming@sz.tsinghua.edu.cnyangelwm@163.com

  • Dr Hantao Liu, Cardiff University, UK

Email: liuh35@cardiff.ac.uk

  • Dr Rui Zhu, University of Kent, UK

Email: r.zhu@kent.ac.uk

  • Prof Xinbo Gao, Xidian University, China

Email: xbgao@mail.xidian.edu.cn

  • Dr Jing-Hao Xue, University College London, UK

Email: jinghao.xue@ucl.ac.uk

Closed Special Issues