Development of a Hairstyle Conversion System Based on Mask R-CNN

Round 1

Reviewer 1 Report

The Paper is well-written, and some of the recommendations are mentioned below:

1. In the abstract, last few lines, authors have mentioned 'our method', it is highly recommended that authors, should change 'our method to 'proposed method' or name the method used in the article or for research work.

2. It has been difficult to understand the terms, hence for the benefit of readers, it is highly recommended authors should add a table of abbreviations.

3. For the understanding of readers, the Authors should explain the results and discussion in a separate section, instead of mixing them in the method section and conclusion section. 

4. Conclusion section should be separate from than discussions. Authors should separate the conclusion section and add future work, and significance of work in the conclusion section too.

Author Response

First of all, thank you very much for your review. We will respond to comments on the paper. We would appreciate it if you could check the part marked in red in the attached file.

Q) 1. In the abstract, last few lines, authors have mentioned 'our method', it is highly recommended that authors, should change 'our method to 'proposed method' or name the method used in the article or for research work.

A) We changed 'our method' to 'proposed method' as the reviewer commented. Please see the attachment.

Q) 2. It has been difficult to understand the terms, hence for the benefit of readers, it is highly recommended authors should add a table of abbreviations.

A) For the convenience of the reader, we have added more terms from other abbreviations rather than adding an abbreviation table. Please see the attachment.

Q) 3. For the understanding of readers, the Authors should explain the results and discussion in a separate section, instead of mixing them in the method section and conclusion section.

A) Conclusions and discussions are divided according to the reviewer's comments to help readers understand without confusing them. Please see the attachment.

Q) 4. Conclusion section should be separate from than discussions. Authors should separate the conclusion section and add future work, and significance of work in the conclusion section too.

A) As in answer 3, the conclusion and discussion sections have been divided and rewritten. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript entitled “Development of hairstyle conversion system based on Mask R-CNN” has been investigated in detail. The topic addressed in the manuscript is potentially interesting and the manuscript contains some practical meanings, however, there are some issues which should be addressed by the authors:

1)      In the first place, I would encourage the authors to extend the abstract more with the key results. As it is, the abstract is a little thin and does not quite convey the interesting results that follow in the main paper. The "Abstract" section can be made much more impressive by highlighting your contributions. The contribution of the study should be explained simply and clearly.

2)      The readability and presentation of the study should be further improved. The paper suffers from language problems.

3)      The “Introduction” section needs a major revision in terms of providing more accurate and informative literature review and the pros and cons of the available approaches and how the proposed method is different comparatively. Also, the motivation and contribution should be stated more clearly.

4)      The importance of the design carried out in this manuscript can be explained better than other important studies published in this field. I recommend the authors to review other recently developed works.

5)      What makes the proposed method suitable for this unique task? What new development to the proposed method have the authors added (compared to the existing approaches)? These points should be clarified.

6)      “Discussion and conclusions” section should be edited in a more highlighting, argumentative way. The author should analysis the reason why the tested results is achieved.

7)      The authors should clearly emphasize the contribution of the study. Please note that the up-to-date of references will contribute to the up-to-date of your manuscript. The study named "Detection of solder paste defects with an optimization‐based deep learning model using image processing techniques" recently published in this journal - can be used to explain the proposed method process in the study or to indicate the contribution in the “Introduction” section.

8)      How to set the parameters of proposed method for better performance?

9)      The complexity of the proposed model and the model parameter uncertainty are not enough mentioned.

10)  The effect of the parametric uncertainty is not discussed in detail. How did the comparison methods perform with or without the uncertainty?

11)  It will be helpful to the readers if some discussions about insight of the main results are added as Remarks.

This study may be proposed for publication if it is addressed in the specified problems.

Author Response

First of all, thank you very much for your review. We will respond to comments on the paper. We would appreciate it if you could check the part marked in red in the attached file.

Q) 1. In the first place, I would encourage the authors to extend the abstract more with the key results. As it is, the abstract is a little thin and does not quite convey the interesting results that follow in the main paper. The "Abstract" section can be made much more impressive by highlighting your contributions. The contribution of the study should be explained simply and clearly.

A) The abstract part was expanded according to the reviewer's comments. Please see the attachment.

Q) 2. The readability and presentation of the study should be further improved. The paper suffers from language problems.

A) To improve readability, English correction was performed and abbreviations were added to some words. Please see the attachment.

Q) 3. The “Introduction” section needs a major revision in terms of providing more accurate and informative literature review and the pros and cons of the available approaches and how the proposed method is different comparatively. Also, the motivation and contribution should be stated more clearly.

A) The introduction part has been expanded by adding different points of the proposed method, motivation and contribution aspects. Please see the attachment.

Q) 4. The importance of the design carried out in this manuscript can be explained better than other important studies published in this field. I recommend the authors to review other recently developed works.

A) We reviewed other recent developments and modified them to add more importance. Please see the attachment.

Q) 5. What makes the proposed method suitable for this unique task? What new development to the proposed method have the authors added (compared to the existing approaches)? These points should be clarified.

A) Revised the paper by clarifying new developments added to the proposed method. Please see the attachment.

Q) 6. “Discussion and conclusions” section should be edited in a more highlighting, argumentative way. The author should analysis the reason why the tested results is achieved.

A) The discussion and conclusion sections have been edited in a controversial way and rewritten into separate sections for discussion and conclusions to analyze the reasons for the test results. Please see the attachment.

Q) 7. The authors should clearly emphasize the contribution of the study. Please note that the up-to-date of references will contribute to the up-to-date of your manuscript. The study named "Detection of solder paste defects with an optimization‐based deep learning model using image processing techniques" recently published in this journal - can be used to explain the proposed method process in the study or to indicate the contribution in the “Introduction” section.

A) In the introduction section, the latter part of the introduction section has been rewritten to emphasize the contribution of the study. Please see the attachment.

Q) 8. How to set the parameters of proposed method for better performance?

A) Other models compared in this paper are models that do not use Mask R-CNN. Because it is a model using StyleGAN, the parameters of other models were not included in the paper, and the parameters of the proposed method were set as the parameters of ResNext-101, which showed the best performance in the backbone network.

Q) 9. The complexity of the proposed model and the model parameter uncertainty are not enough mentioned.

A) Unlike other models, the model proposed in this paper describes a model that uses Mask R-CNN to improve GAN performance, so only the parameters of the network used in Mask R-CNN are inserted into the table. Since we confirmed that uncertainty such as epistemic uncertainty is the most suitable model through the Mask R-CNN paper, we omitted the description of the uncertainty by referring to the Mask R-CNN paper.

Q) 10. The effect of the parametric uncertainty is not discussed in detail. How did the comparison methods perform with or without the uncertainty?

A) In this paper, we propose a model that improves the performance of GAN with Mask R-CNN, so we confirm a comparative method with or without uncertainty through Mask R-CNN paper and are not affected by model parameter uncertainty. As with answer 9, we solved the part about uncertainty in Mask R-CNN, and we did not discuss the effect of uncertainty on parameters because we proposed a GAN model with improved performance using Mask R-CNN.

Q) 11. It will be helpful to the readers if some discussions about insight of the main results are added as Remarks.

A) According to the reviewer's comments, the previously integrated discussion and conclusion were rewritten by dividing it into discussion and conclusion sections. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The Paper is well improved and all comments are addressed appropriately. I have a minor comment for the authors to please add a table of abbreviations for readers to easily relate to different terms in the article.

Author Response

Thank you very much for your comment. I added comments from reviewers to create an appendix and added an abbreviation table. Add a modified version.

Author Response File: Author Response.pdf

Reviewer 2 Report

All my comments have been thoroughly addressed. It is acceptable in the present form.

Author Response

Thank you for your comment. I feel very grateful to the reviewer.

Thank you very much.