Applied Research on Face Image Beautification Based on a Generative Adversarial Network
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsTitle: Applied Research on Face Image Beautification Based on a 2 Generative Adversarial Network
The work presented by the authors is of interest, but there are some important points that need to be improved. My comments are as follows.
What are the numbers in almost all keywords?
Much of the text from the "Introduction" section should be moved to the "Related Work" section, particularly from "Peng et. al..." (line 42) to "... features or Asian faces." (line 83). The "Related Work" section should be restructured to integrate this text in a consistent and consistent manner. As a result of this restructuring, the "Introduction" section should be supplemented with updated and concise information.
The literature review should be done not only analytically but also critically. For example, the text written in subsection 2.1 is a list of works by other authors, without any connection, requiring analysis and discussion.
Line 175 - In "... random vector to the style vector..." seems that something is missing.
Line 196 - "... input into the encoder..." needs revision. Same comment for similar situations, like the one in line 202. The same in lines 225, 226 and 227.
When using an acronym for the first time, its name must be written out in full (for example FPN, on the line 207).
In general, the quality of the images should be improved (in particular the style of the text).
Lines 242 and 243 - Text must be revised.
Equations should be referred in the text. For instance, in line 304, write "... is expressed by equation 1" instead of "... is expressed as"
Between each section and subsection titles should be written at least one paragraph.
The titles of subsections 4.1.1 e 4.1.2 are the same.
The authors state that the choice of image resolution is made in order to balance computational resources, but nothing is said about the computational cost of the proposed approach. This study is important for a more rigorous/in-depth performance evaluation.
Lines 418 to 420 - The claim should be justified. What are the visual differences that are in the basis of such conclusion?
Conclusions should begin with a brief contextualization of the work presented in the paper.
What are the limitations of the presented study?
Comments on the Quality of English LanguageThe manuscript must be carefully reviewed, eliminating typos (mostly grammatical) and improving sentence constructions.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsNeed changings.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsDear Authors,
Thank you for reviewing the initial manuscript, taking into account my comments and suggestions.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper improved
Comments on the Quality of English LanguageThe English could be improved to more clearly express the research.