Lightweight Facial Expression Recognition Based on Class-Rebalancing Fusion Cumulative Learning

Round 1

Reviewer 1 Report

The manuscript is well written. My few suggestions to improve the manuscripts are: Pg9ln318 - Weight sharing in the dual-branch network: semicolon instead of period. Similarly, Ph9ln324 – the first phrase should have a semicolon instead of a period. Pg90 &10 ln365 to 384 – describing the retrieved datasets – please provide the reference or URL so the readers are aware of the datasets extracted for this study. Since the authors have mentioned that this is an ongoing study, it will be helpful at some point to share their data with the scientific community so it can be further validated.

Author Response

Dear Editors and Reviewers:

    Thanks for your letter and for reviewer’s comments concern our manuscript. Those comments are valuable and helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied all comments carefully and have made correction which we hope meet with approval. Revised portion are marked with underline in the paper.

Thank you and best regards.

Yours sincerely,
Xiangwei MOU

Author Response File: Author Response.pdf

Reviewer 2 Report

The article's topic concerns very significant and up-to-date issue of Facial Expression Recognition. The aim of Authors is to propose a lightweight and efficient method to improve recognition accuracy.

The background of the research and its significance is explained in a very proper manner and the main contribution of Authors is clearly explained. The method proposed by the Authors is tested in the experiment and compared with other methods. 

The references that are used are appropriate and provide support to the Authors claims nad conclusions. 

All the figures and tables provide significant information that expands the text of article.

I assess the value of the paper as high and recommend to publish it without any revision.

Author Response

Dear Editors and Reviewers:

    Thanks for your letter and for reviewer’s comments concern our manuscript. Those comments are valuable and helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied all comments carefully and have made correction which we hope meet with approval. Revised portion are marked with underline in the paper.

Thank you and best regards.

Yours sincerely,
Xiangwei MOU

Author Response File: Author Response.pdf

Reviewer 3 Report

The research adresse an interesting question, what is the potential of new advancements in research of Facial Expression Recognition?

I consider the article relevant in the field because a detailed knowledge of convolutional neural networks. The paper provides an update on the main current and future imaging techniques but also a dual-branch network is proposed .

There are no specific improvements to be done.

The conclusions are consistent with the purpose of this research in order to provide a solution for the problem in recognition accuracy model.

The references are among the newest and most relevant in the field and the paper comes with a rich paraclinical imagery that completes the information presented in the text.

Author Response

Dear Editors and Reviewers:

    Thanks for your letter and for reviewer’s comments concern our manuscript. Those comments are valuable and helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied all comments carefully and have made correction which we hope meet with approval. Revised portion are marked with underline in the paper.

Thank you and best regards.

Yours sincerely,
Xiangwei MOU

Author Response File: Author Response.pdf

Reviewer 4 Report

Reviewer understands that Mou et al. has presented a manuscript entitled "Lightweight Facial Expression Recognition Based on Class-Rebalancing Fusion Cumulative Learning". Reviewer has a few suggestions, and they request that the authors kindly answer all the questions by updating the requested details in their manuscript.  

1) Sub-details from Fig. 7, Fig. 8 and Fig. 9 are non-readable. Make sure that each and every sub-detail is readable in A4/letter size print. Provide large sized HD figures.

2) On page 12, on lines 116, 117, 119 and 121 100th, 200th, 150th and 180th is there. Please write "th" in the superscript format.

3) Discuss the repeatability and reusability your used method.

4) Please mention your study’s limiting or impacting factors/ parameters at previous stages and how you worked on them in your presented studies.

5) In section 5: Conclusions, please mention about current limitations of your work and the future scope of improvements.

 

Author Response

Dear Editors and Reviewers:

    Thanks for your letter and for reviewer’s comments concern our manuscript. Those comments are valuable and helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied all comments carefully and have made correction which we hope meet with approval. Revised portion are marked with underline in the paper.

Thank you and best regards.

Yours sincerely,
Xiangwei MOU

Author Response File: Author Response.pdf