A Hybrid Contrast and Texture Masking Model to Boost High Efficiency Video Coding Perceptual Rate-Distortion Performance

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper designs two techniques, i.e. contrast masking, and texture masking, to boost HEVC perceptual RD performance. Although valuable experimental results are obtained, some issues should be addressed.

1. How to calculate the proposed 4X4 weighting matrices in Figure 3?

2. The SVM model is implemented with MATLAB R2020a, how to integrate it into the HM? Why not use the C version of the SVM model?

3. The authors mention luminance masking in the section of the introduction but do not use it in the following text. Why?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Authors of this paper present a scheme combining contrast and texture masking techniques for the High Efficiency Video Coding (HEVC) video codec. The proposed concept has been tested in terms of objective metrics. The content of the article meets with the topics of Electronics journal and also meets with the topics of SI.

Overall, the article has a united concept. However, it needs major revision.

 

Notes:

o   Abstract – the paper abstract should better capture the proposed technique's main novelty.

o   Article – please, check that all the used abbreviations are explained (e.g., DCT).

o   Article - there are applied two ways to explain abbreviations: “Human Visual System (HVS)” and “QP (Quantization Parameter)”. Please, do not mix them, use only one of them. Thanks!

o   Section 2 – “PSNR” should be defined first (line 98); (and not in line 124).  

o   Introduction / Section 2 – the state-of-the-art (SoTA) is evaluated on good level. However, authors should work with more recent articles, mainly published in the period of 2020-2024. The authors should consider about the extension of SoTA with an appropriate works, for instance with “Subjective Quality Assessment of the Impact of Buffer Size in Fine-Grain Parallel Video Encoding” – in which the authors explore the impact of buffer size on image subjective quality, and its relation with other encoding parameters, and with “Object Tracking Based Surgical Incision Region Encoding using Scalable High Efficiency Video Coding for Surgical Telementoring Applications” providing interesting study of the mix of HEVC codec and machine learning approach to improve bit rate versus visual quality in the surgical telementoring systems. I hope that the mentioned work can be helpful in the improvement of the elaboration of SoTA and can give helpful information for potential readers. Otherwise, please, work with other paper(s). Thank you! Next, please, explain the main differences between this and your previous works.

o   Introduction / Section 2 – the paragraph dealing with the organization of the paper should be presented at the end of the Introduction.

o   Section 2 – I recommend for the authors to more highlight the main contributions of the paper – present them by items in separated lines.

o   Section 3 – Fig. 2 – there are two curves but the legend in the figure is missing.

o   Section 3 – Table 1 – snapshots of the vide sequences as well as their short description is missing.

o   Section 3 – Fig. 4 (a)-(b) – there is a video in Full HD. The fourth bitrate value is around 130 and 160 Mbps. Are these values so high after the compression versus the values of objective metrics?

o   Section 3 – there are typos – “USC-SIPI Image Database [?]”; Kodak image dataset [?].

o   Article – all the used HW/SW equipment/configuration must be presented in detail (please, check it). Next, for reproducible research, I would suggest the authors make the source codes publicly available.   

o   Section 3 – the visual quality of some figures, like Figs. 9-10, should be improved (they are slightly blurred).

o   Sections 3 and 4 – the performance of the proposed solutions should be better compared with SoTA solutions. Next, the author should consider more (and mainly advanced) objective metrics (not only PSNR and SSIM-based ones).

Section 4 – “After performing a subjective analysis it is quite difficult to see any difference between the two pictures.” Why the authors do not provide real results from subjective test conducted under controlled laboratory conditions?

Comments on the Quality of English Language

See "Comments and Suggestions for Authors".

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed my questions, So I suggest to accept it for publication.

Author Response

Thank you for your positive feedback and recommendation for publication. We greatly appreciate your support.

Reviewer 2 Report

Comments and Suggestions for Authors

The article has been partly improved. Many thanks for the explanation letter! Next time, please, indicate in the explanation letter the changes (name of section or line number) provided in the article in a more visible form. Thanks!

After the check of the article, I have the following notes.

Notes:

-     Figure 2 – for some readers, abbreviation “cpd” should be misleading. Please, provide its full name in the caption of the figure.

-        Article – if the source code of the application is not currently available, then the authors of the paper must provide information in the article about that the source code, or the SW application will be available for a reader on request.

-        Figure 8 – please, explain in detail that why you have some many outlier values in the presented box plots.

-        Section 4 – “The system is equipped with 376 GB of RAM.” Really, 376 GB of RAM?

Article / Explanation Letter – “To address your suggestion, we will explore the inclusion of additional advanced objective metrics such as VMAF or VIF.” I do not see any changes in the article from this point of view. As a I mentioned, it should be nice to complete the provided study with scores from other (relevant and advanced) objective metrics.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

The article has been improved.