Adaptive Content Frame Skipping for Wyner–Ziv-Based Light Field Image Compression

Round 1

Reviewer 1 Report

1) The authors consider other DVC or traditional video coding standards, i.e., H.264/H.265, as a main competitors for the proposed codec. But there are other prediction-free codecs, which could be used. For example, video codecs based on three-dimensional discrete wavelet transform (3-D DWT [a]) or video coding based on compressive sensing (CS-JPEG [b]). I think the authors should at least mention the alternative coding possibilities in Introduction.

[a] E.Belyaev, K.Egiazarian and M.Gabbouj, A low-complexity bit-plane entropy coding and rate control for 3-D DWT based video coding // IEEE Transactions on Multimedia, vol.15, iss.8, pp.1786 – 1799, 2013.

[b] E.Belyaev, Compressive Sensed Video Coding having JPEG compatibility // IEEE International Conference on Image Processing (ICIP), 2020.

2) It seems that the feedback channel requires repetition of heavy computations at the decoder side. Moreover, such decoding should be performed in real-time, which could be not possible in many applications. Is there any rate control considered within the work?

3) It is well known that the scanning order based on minimum order [c] provides the best performance, but it is not considered in this work. Please consider this order as well.

[c] J. S. Ricardo Monteiro et al. "Optimized Reference Picture Selection for Light Field Image Coding," EUSIPCO, 2019.

4) It seems that the decision tree based adaptive frame skipping requires many computations per pixel including log, i.e., this complexity could be even higher than initial WZ frame encoding complexity. I think it could be nice to add complexity analysis with and without skipping.

5) Usually, H.265/HEVC Inter No Motion (i.e., without motion compensation) is also considered within the comparison of DVC and traditional codecs. I think it makes sense to add such comparison as well.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

In this manuscript, the authors propose a new LF coding architecture based on WZ approach.

The manuscript is well-structured.

The research motivation, goal and contributions are clearly formulated.

The experimental results demonstrate that the new adaptive frame skipping algorithm has better compression performance in comparison with state-of-the-art video compression standards.

There is no information on how the new method is actually implemented (programming language, development environment).

The reviewer’s main concern is the following:

The authors needs to explain their algorithm by using relevant step-by-step explanations. Currently, there are only several lines written in pseudocode and it is not easy for readers to understand and reproduce the results.

 

Some technical remarks:

l. 277: The colours of two blue series are too similar.

l. 296: In Eq. (2) variable D is undefined and an index in summation is missing. Actually, the text in lines 285-303 in not related to the rest of subsection 3.2., and could be omitted.

l. 313, 315, 323: there is some confusion in notations used in description and those in equations, for example SAD_SAI or SAI_SAD, etc.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

In this paper, the authors propose a Light Field image content-adaptive frame skipping compression solution to achieve high compression performance. In doing so, they develop a decision tree is known as a classification rule or a set of if-then rules that determines the class of any object from the values of its attributes. In the implementation process, a decision tree is known as a classification rule or a set of if-then rules that determines the class of any object from the values of its attributes is utilized.

Moreover, the authors employ the concepts of entropy work.  This paper has shown how they apart from that the paper proposes a business level.

In experimental results, the paper contains of material to show the validity of the proposed method

Author Response

Thanks the reviewer for the positive summarization.

Round 2

Reviewer 2 Report

The quality of the revised electronics-962203
“Content Adaptive Frame Skipping for Wyner-Ziv based Light Field Image Compression” manuscript
has been substantially improved.
The overall presentation of the proposed light field coding
architecture based on Wyner-Ziv approach is corresponding to Electronics
journal publishing standard.
My recommendation is to accept the manuscript for publication.