Robust RPPG Method Based on Reference Signal Envelope to Improve Wave Morphology

Round 1

Reviewer 1 Report

The manuscript entitled “Robust RPPG Method Based on Reference Signal Envelope to Improve Wave Morphology” has been investigated in detail.
The paper’s subject could be interesting for readers of journal. Therefore, I recommend this paper for publication in this journal but before that, I have a few comments on the text that should be addressed before publication:

 

·         Some abbreviations are missing in the text, please check the manuscript thoroughly to find the ones and add them in the table.

·         Please mention clearly in the text that what is the main benefit of using remote photoplethysmography (RPPG) in this study.

·         Authors should clearly write what the motivation of this paper is. Why specifically … algorithm is used for the … problem is not clear.

·         How did you evaluate the accuracy of obtained results?

·         The linguistic quality needs improvement. It is essential to make sure that the manuscript reads smoothly- this definitely helps the reader fully appreciate your research findings. There are grammar and writing style errors that should be corrected by the authors.

·         Some equations should be used with correct citations. They seem as if they are proposed and used firstly in this paper.

·         Mentioned errors in the conclusion section must be compared with the previous studies in the field

·         Some more recommendations and conclusions should be discussed about the paper considering the results. The Conclusion section is weak

Author Response

Point 1: Some abbreviations are missing in the text, please check the manuscript thoroughly to find the ones and add them in the table.

Response 1: We are very sorry for our negligence. We have added abbreviations in and after the text, and a brief explanation in the text when it first appeared.

Point 2: Please mention clearly in the text that what is the main benefit of using remote photoplethysmography (RPPG) in this study.

Response 2: Thank you very much for the reminding. We have adjusted the content of the introduction to include recent literature about the benefits and applications of RPPG (especially in the medical field) (line 54-67 in new manuscript). We also added evidence about using PPG to improve RPPG signals and described our motivation for proposing the algorithm in the context of modern health monitoring (line 68-85 in new manuscript).

Point 3: Authors should clearly write what the motivation of this paper is. Why specifically … algorithm is used for the … problem is not clear.

Response 3: Thanks for the kindly comment. As described in the previous point, we have added a description of motivation to the introduction, supported by references to recent relevant research (line 68-85 in new manuscript).

Point 4: How did you evaluate the accuracy of obtained results?

Response 4: Thank you for pointing out the problem. Sorry for that I am not fully understanding the question. In my opinion, we repeated the experiments on 12 subjects (11 valid) and synthesized the results of them to conduct ANOVA. From the statistical point of view, the proposed method has a significant advantage in envelope approximation. In addition, the signal curve is visually analyzed. As mentioned in the Discussion section, the proposed method makes morphological improvements to the G channel. However, the experiment is also limited to the number and type of subjects, which is one of the shortcomings of the research. Future study will recruit more diverse subjects.

Point 5: The linguistic quality needs improvement. It is essential to make sure that the manuscript reads smoothly- this definitely helps the reader fully appreciate your research findings. There are grammar and writing style errors that should be corrected by the authors.

Response 5: Thanks for the comments and we are sorry for the mistakes. We have thoroughly checked and corrected the grammar and writing style errors we found in our revised manuscript. Hope that readability is improved.

Point 6: Some equations should be used with correct citations. They seem as if they are proposed and used firstly in this paper.

Response 6: Thanks for your suggestion. We added the citation to the original equation according to the comments, and the rest of the equations are based on its expansion or what we came up with.

Point 7: Mentioned errors in the conclusion section must be compared with the previous studies in the field.

Response 7: Thank you for the consideration. Sorry for that we didn't fully understand "errors". Actually, we have fleshed out the conclusions, including a brief background (and related recent work), the content of this study, and responses to the results and discussion. The limitations of the research are placed in the Discussion section.

Point 8: Some more recommendations and conclusions should be discussed about the paper considering the results. The Conclusion section is weak.

Response 8: Thanks a lot for the concern. As mentioned in the previous point, we have strengthened the conclusions and expanded the content according to your comments.

Reviewer 2 Report

In this article, the authors presented a method that enhances RPPG signals using an envelope-based approach, effectively preserving pulse information while reducing noise. The investigation into the envelope feature of PPG signals showcases significant progress. However, I have a few questions and suggestions that need to be addressed before considering the manuscript for further review.

 

1.You mentioned decomposing the RPPG signal into three components: pulsation, constant, and disturbance. After this decomposition, it would be helpful to know what steps you took to reduce noise from these components.

2.It would be beneficial to provide a brief explanation of RGB (Red, Green, Blue) when you first introduce the abbreviation, instead of using the abbreviation directly.

3.In line 151, you refer to "x(t)" - does this mean "xc(t)"? Additionally, in line 167, you mention a "standard y(t)," but I did not find any equation referring to y(t). Do you mean y(t) = 1*xG(t)+... in equation 2?

4.Please provide an explanation for the "length parameter" and its effect on smoothing. Currently, the length parameter "n" is not referenced in any equation, which may confuse readers.

5.I suggest combining Figure 2 and Figure 3, as only the middle plot in Figure 3 appears to be new in comparison to Figure 2. To enhance clarity, you can present the combined figures from top to bottom as follows: G channel, combined signal, smoothed combined signal, and finger PPG curve.

6.Could you please explain the difference between the groundtruth finger PPG signal and the signals from the GRB channel? Additionally, where does the groundtruth finger PPG signal come from? Was it obtained using the Logitech C1000e camera as well?

7.In line 272, you mention that "Higher fill light intensity leads to significantly better light information of finger skin." Please provide a more contextual explanation for this statement. When comparing "weak fill" and "strong fill," the G and B channels of the finger skin exhibit higher standard deviation, indicating less stability. Only the R channel of the finger skin shows a lower standard deviation. What is meant by "significantly better light information"? Clarifying this statement within the appropriate context would be beneficial.

8.In Figure 9, it would be sufficient to include only one of the subjects, while providing the remaining three subjects' data as supplementary information. This will allow readers to view the lines, x-axis, and y-axis more clearly. The same approach can be applied to Figures 10, 11, and 12.

9.In line 321, you mention the "benchmark PPG signal." Please clarify where this benchmark PPG signal comes from. Does the mean value indicate of the benchmark PPG signal?

Addressing these questions and suggestions will significantly improve the clarity and comprehensibility of your manuscript. We look forward to reviewing the revised version of your paper.

Author Response

Point 1: You mentioned decomposing the RPPG signal into three components: pulsation, constant, and disturbance. After this decomposition, it would be helpful to know what steps you took to reduce noise from these components.

Response 1: Thanks for pointing out the issue and we are sorry for not describing it clearly. We adjusted the equations (1 and 2 in new manuscript) and filled in the details. Equation 1 describes the RPPG decomposition at a single pixel, including pulsation, constant, and disturbance noise. Equation 2 is the result of equation 1 averaging over the pixels of ROI, where ordinary white noise (nC) vanishes. Ambient light effects, motion artifact, light absorption and reflection of bone and muscle are all grouped into the constant components of non-pulsation (a_d,C). In fact, we only eliminate the system white noise by averaging the pixels, and the other noise is not processed in the equation segment.

Point 2: It would be beneficial to provide a brief explanation of RGB (Red, Green, Blue) when you first introduce the abbreviation, instead of using the abbreviation directly.

Response 2: Thanks for the reminding and we are sorry for the negligence. We have added abbreviations in and after the text, and a brief explanation in the text when it first appeared.

Point 3: In line 151, you refer to "x(t)" - does this mean "xc(t)"? Additionally, in line 167, you mention a "standard y(t)," but I did not find any equation referring to y(t). Do you mean y(t) = 1*xG(t)+... in equation 2?

Response 3: Thank you so much for your careful check. We are sorry about the mistake. We have corrected and adjusted the equations. Actually, "xc(t)" is one channel signal and "x(t)" is a combined signal of RGB channels. A "standard y(t)" in old line 167 (new line 233) comes from the text in old line 148 “We suppose the finger PPG as the groundtruth, which provides a pulse signal with an accurate envelope curve y(t)” (new line 202).

Point 4: Please provide an explanation for the "length parameter" and its effect on smoothing. Currently, the length parameter "n" is not referenced in any equation, which may confuse readers.

Response 4: We are sorry about the confusion. To improve readability and avoid confusion, we deleted “n” in line 256 in new manuscript. Actually, during the experiment, we found that either too long or too short length was inappropriate, so we chose length for 10-second in the article (new line 256-258).

Point 5: I suggest combining Figure 2 and Figure 3, as only the middle plot in Figure 3 appears to be new in comparison to Figure 2. To enhance clarity, you can present the combined figures from top to bottom as follows: G channel, combined signal, smoothed combined signal, and finger PPG curve.

Response 5: We fully agree with your suggestions. We have combined the four curves into one figure (Figure 2 in new manuscript).

Point 6: Could you please explain the difference between the groundtruth finger PPG signal and the signals from the GRB channel? Additionally, where does the groundtruth finger PPG signal come from? Was it obtained using the Logitech C1000e camera as well?

Response 6: Thanks for your concern. The groundtruth PPG signal comes from the finger PPG device in line 276-285 in new manuscript. And the RGB channels is collected by Logitech C1000e camera.

Point 7: In line 272, you mention that "Higher fill light intensity leads to significantly better light information of finger skin." Please provide a more contextual explanation for this statement. When comparing "weak fill" and "strong fill," the G and B channels of the finger skin exhibit higher standard deviation, indicating less stability. Only the R channel of the finger skin shows a lower standard deviation. What is meant by "significantly better light information"? Clarifying this statement within the appropriate context would be beneficial.

Response 7: Thanks a lot for your careful check. We rephrased it more appropriately and contextually in line 345-348 in new manuscript.

Point 8: In Figure 9, it would be sufficient to include only one of the subjects, while providing the remaining three subjects' data as supplementary information. This will allow readers to view the lines, x-axis, and y-axis more clearly. The same approach can be applied to Figures 10, 11, and 12.

Response 8: Thank you for the suggestion. We have adjusted the content of the picture. Only one subject is included in the paper, and the data of the other three subjects are included in the Appendix.

Point 9: In line 321, you mention the "benchmark PPG signal." Please clarify where this benchmark PPG signal comes from. Does the mean value indicate of the benchmark PPG signal?

Response 9: Thanks for the comments. In this paper, we have three equivalent statements for finger PPG signals: reference, groundtruth, and benchmark signal. The "benchmark PPG signal" actually represents the original signal collected by the finger PPG device.

Reviewer 3 Report

Reviewer Comments

 

Manuscript Title: Robust RPPG Method Based on Reference Signal Envelope to

Improve Wave Morphology

Manuscript Number: electronics-2434485

 

The manuscript under review is devoted to studying an envelope-based method for remote photoplethysmography channels to improve the wave morphology of the collected signal based on the reference signal from finger Photoplethysmography. Using a model consistent with physiological and optical principles, the authors divided the signal into a linear superposition of pulse, constant, and disturbance components. The correlation coefficients were used to calculate a linear combination of RGB channels so that the envelope shape of the combined signal was similar to that of the benchmark signal. The authors designed experiments with different intensities and stability of illumination, to test the envelope approximation ability and robustness of the proposed method and some commonly used methods.

The manuscript contains new and significant. The abstract clearly and accurately describes the content of the article. The literature review part contains distinct and rich references. The paper is nicely written and can be accepted but first, it should be improved. I have these comments:

1- In line 146, authors must demonstrate their choice of the MATLAB function ’envelope’ used to obtain the signal envelope in the experiment.

2- Line 185, authors mention that the parameters as, C(t) = as, C, ad, C(t) = ad, C vary from person to person and can be affected by skin color, age, and gender. How you can define these constants and which values are used during the experiment?

3- In equation 1, how do you determine the value of the noise nC(t)?

4- In Figure 2, and Figure 3, it is best to add a subtitle to each graph and label them (a), (b), etc.

Finally, I recommend the paper for publication after resolving the minor comments.

 

 

Author Response

Point 1: In line 146, authors must demonstrate their choice of the MATLAB function ’envelope’ used to obtain the signal envelope in the experiment.

Response 1: Thank you for pointing out the issue. We have added the parameter information in the text (line 199 in new manuscript).

Point 2: Line 185, authors mention that the parameters as, C(t) = as, C, ad, C(t) = ad, C vary from person to person and can be affected by skin color, age, and gender. How you can define these constants and which values are used during the experiment?

Response 2: Thanks a lot for the concern. In this study, the specific values of these constants do not affect the results. The function of the constants is to support the rationality of the linear combination of RGB channels (equation 3 in new manuscript). However, we are also interested in the specific constant values, which may be taken into account in future studies.

Point 3: In equation 1, how do you determine the value of the noise nC(t)?

Response 3: We are sorry for the confusion. We have adjusted the equations (1 and 2 in new manuscript) and filled in some details. Equation 1 describes the RPPG decomposition at a single pixel, including pulsation, constant, and disturbance noise. Equation 2 is the result of equation 1 averaging over the pixels of ROI, where ordinary white noise (nC) vanishes. Ambient light effects, motion artifact, light absorption and reflection of bone and muscle are all grouped into the constant components of non-pulsation (a_d,C). In fact, we only eliminate the system white noise by averaging the pixels, and the other noise is not processed in the equation segment.

Point 4: In Figure 2, and Figure 3, it is best to add a subtitle to each graph and label them (a), (b), etc.

Response 4: Thank you for the consideration. We have combined the four curves into one figure and added subtitles (Figure 2 in new manuscript).

Reviewer 4 Report

 

The Article Robust RPPG Method Based on Reference Signal Envelope to Improve Wave Morphology, seems to cover a hot topic, however, the authors include a list of very old references. The newest one is from 2012, two years ago:

15. Kuwałek, P.; Burlaga, B.; Jesko, W.; Konieczka, P. Research on methods for detecting respiratory rate from photoplethysmographic 429

signal. Biomed. Signal Process. Control. 2021, 66, 102483. Authors must search related and very recent Works to describe the state of the art and then to highlight the main contribution.

The section 6. Conclusions is too short to summarizy main findings of this work, and the mentioned methods need to be justified by including their related references, which must be recent to appreciate the main advances in the recent couple of years. In fact, the authors emphasize some methods that need to be documented in the manuscript, see the related sentences in the conclusion section:

In this study, we present a novel method for enhancing RPPG signals by combining RGB channels with groundtruth signals from finger PPG… The novel Word must mean that it  is compared to recent Works published in the last couple of years… Our proposed EB method outperforms the G-R, G-B, G, CHROM, and POS methods under various lighting conditions, providing a stable and efficient approximation effect while also improving the morphology of the envelope… To lead to this conclusion, authors must should compare the method with recen tones and list the most recent and related references in each case… We also discuss the impact of environmental stability on each method. Our findings demonstrate the effectiveness of the proposed method and its potential to enhance the accuracy of RPPG signals in various scenarios… This final conclusion again requires a revision of very recent Works that must summarize main advantages in a Table, if possible.

 

The plots given from Figure 9. Representative waveforms under ’weak fill’ light condition, from four subjects. The left picture of each subgraph shows, from top to bottom, the raw signals (blue lines) of EB, CHROM, POS, and finger PPG, and their upper envelopes (orange lines), respectively; on the right side of the subfigure, raw signals of the G-channel, EM method, and finger PPG are presented from top to

bottom… to Figure 12, can be summarized in a Table or in boxes as done in Figure 8. Box-plots of envelope correlation coefficients for multiple methods under four illumination conditions: (a) weak fill light; (b) strong fill light; (c) LED lamps; (d) LED lamps and strong fill light. The methods represented from left to right in each small figure are EB, G-R, G-B, G, CHROM, and POS. Interquartile ranges are indicated by boxes, median values by horizontal lines inside the boxes, mean values by crosses, and outliers by circles.

 

The analysis of variance is good, however, it requires justification of other measurements and to include realted references to perform a better comparison to enhance Table 1. Results of ANOVA between EB and other methods. When the p-value was less than 0.05, the two methods were considered to be significantly different.

 

Section 3.2.4. Compared methods We compared the EB method with commonly accepted CHROM [11 ] and POS [12 ] methods, as well as G [9], G-R [ 10] and G-B (green channel minus blue channel) methods…. These sections include old references that may not be enought to appreciate the advance in the state of the art.

minor revision

Author Response

Point 1: The Article Robust RPPG Method Based on Reference Signal Envelope to Improve Wave Morphology, seems to cover a hot topic, however, the authors include a list of very old references. The newest one is from 2012, two years ago: ‘15. Kuwałek, P.; Burlaga, B.; Jesko, W.; Konieczka, P. Research on methods for detecting respiratory rate from photoplethysmographic signal. Biomed. Signal Process. Control. 2021, 66, 102483.’ Authors must search related and very recent Works to describe the state of the art and then to highlight the main contribution.

Response 1: Thank you for the suggestions. We have adjusted the content of the introduction to include recent literature about the benefits and applications of RPPG (especially in the medical field) (line 54-67 in new manuscript). We also added evidence about using PPG to improve RPPG signals and described our motivation for proposing the algorithm in the context of modern health monitoring (line 68-85 in new manuscript).

Point 2: The section 6. Conclusions is too short to summarize main findings of this work, and the mentioned methods need to be justified by including their related references, which must be recent to appreciate the main advances in the recent couple of years. In fact, the authors emphasize some methods that need to be documented in the manuscript, see the related sentences in the conclusion section:

‘’In this study, we present a novel method for enhancing RPPG signals by combining RGB channels with groundtruth signals from finger PPG… The novel Word must mean that it is compared to recent Works published in the last couple of years… Our proposed EB method outperforms the G-R, G-B, G, CHROM, and POS methods under various lighting conditions, providing a stable and efficient approximation effect while also improving the morphology of the envelope… To lead to this conclusion, authors must should compare the method with recent tones and list the most recent and related references in each case… We also discuss the impact of environmental stability on each method. Our findings demonstrate the effectiveness of the proposed method and its potential to enhance the accuracy of RPPG signals in various scenarios… This final conclusion again requires a revision of very recent Works that must summarize main advantages in a Table, if possible.’’

Response 2: Thank you very much for pointing out the issues. We have fleshed out the conclusions, including a brief background (and related recent work), the content of this study, and responses to the results and discussion.

We are sorry for not including the most recent references. But as mentioned in the introduction, RPPG signals are often used to target physiological variables, such as heart rate. Measurement methods are divided into model analysis and data-driven deep learning methods, of which the latter is far more popular than the former in recent years. However, in order to achieve more complex and diverse health monitoring, the model analysis of RPPG signals is important, although there has been little work in recent years. The purpose of this paper is to improve the RPPG signal from a fundamental model perspective, so the references may not be sufficiently new.

Point 3: The plots given from Figure 9. Representative waveforms under ’weak fill’ light condition, from four subjects. The left picture of each subgraph shows, from top to bottom, the raw signals (blue lines) of EB, CHROM, POS, and finger PPG, and their upper envelopes (orange lines), respectively; on the right side of the subfigure, raw signals of the G-channel, EM method, and finger PPG are presented from top to bottom… to Figure 12, can be summarized in a Table or in boxes as done in Figure 8. Box-plots of envelope correlation coefficients for multiple methods under four illumination conditions: (a) weak fill light; (b) strong fill light; (c) LED lamps; (d) LED lamps and strong fill light. The methods represented from left to right in each small figure are EB, G-R, G-B, G, CHROM, and POS. Interquartile ranges are indicated by boxes, median values by horizontal lines inside the boxes, mean values by crosses, and outliers by circles.

Response 3: Thanks for your comments. To improve readability, we have adjusted the content of the picture. Only one subject is included in the paper, and the data of the other three subjects are included in the Appendix. Considering that an image may have multiple features, we ultimately decided to retain the original intuitive figures without adding a tabular description. The morphological improvements to G channel by the proposed method are described and summarized in Section 5.2.

Point 4: The analysis of variance is good, however, it requires justification of other measurements and to include realted references to perform a better comparison to enhance Table 1. Results of ANOVA between EB and other methods. When the p-value was less than 0.05, the two methods were considered to be significantly different.

Response 4: Thanks a lot for the consideration. Two measures were considered in this study, a statistical significance test and an intuitive degree of improvement. Since the target of this paper is the approximation of envelope shape, we use correlation coefficient to show the fitting ability of each method. Therefore, only ANOVA was performed for correlation coefficients. In order to further explain the envelope approximation degree of different methods, we choose the signal curves of representative subjects to give a more direct and clear explanation.

Point 5: Section 3.2.4. Compared methods We compared the EB method with commonly accepted CHROM [11 ] and POS [12 ] methods, as well as G [9], G-R [ 10] and G-B (green channel minus blue channel) methods…. These sections include old references that may not be enought to appreciate the advance in the state of the art.

Response 5: Thanks a lot for your concern. As in Response 2, we expect to improve the RPPG signal from a fundamental model perspective, but recent work was limited. And we are really sorry for not including the references in the state of the art. We have added some recent references in the introduction.

Reviewer 5 Report

please see the attachment 

Comments for author File: Comments.pdf

Manuscript needs a substantial review for grammar and style. There are awkward sentences and typographical errors which, although minor, contribute to the manuscript's overall lack of clarity. 

Author Response

Point 1: The abstract states that "The results showed that the envelope-based approach performed better in most cases". However, it could benefit from more specific details about the results. How much better did it perform? Were there any metrics used to measure this? Similarly, was there a significant improvement in the waveform of the original G channel?

Response 1: Thank you for the consideration. We have adjusted the content of the abstract to describe in more detail how the proposed method performs better.

Point 2: The abstract has minor grammatical issues. For example, "Remote physiological monitoring has becoming increasingly important" should be corrected to "Remote physiological monitoring has become increasingly important". Also, "we divided the signal into linear superpositions of pulse, constant, and disturbance components", might better be phrased as "we divided the signal into linear superpositions comprising pulse, constant, and disturbance components".

Response 2: Thanks for the comments and we are sorry for the mistakes. We have checked and corrected the grammar and writing style errors we found in the abstract.

Point 3: The abstract uses several abbreviations without first introducing them, such as "PPG" and "RGB". It's usually good practice to spell out abbreviations when they first appear in the text to avoid confusion for readers unfamiliar with the specific terminology.

Response 3: Thank you so much for your careful check. We have added abbreviations in and after the text (including abstract), and a brief explanation in the text when it first appeared.

Section 1

Point 4: There are a few grammatical errors. For example, "and has various applications" should be "and have various applications". Also, "This study focuse on using a" should be "This study focuses on using a".

Response 4: Thanks for the comments and we are sorry for the mistakes. We have thoroughly checked and corrected the grammatical errors we found in our revised manuscript. Hope that readability is improved.

Point 5: The readability can be improved by splitting some of the longer sentences into shorter ones.

Response 5: Thank you for your suggestion. We adjusted the overly complex sentence structure in the article, as well as some inappropriate words. Hope that readability is improved.

Point 6: The explanation of the PPG signals, the concept of the 'envelope', and the methods to address environmental interference are well laid out. However, for readers unfamiliar with the field, it might be beneficial to provide a brief definition or explanation of certain terms like "RGB channels", "G", "G-R", "CHROM", and "POS".

Response 6: Thanks for your valuable comments. We have added abbreviations in and after the text (including abstract), and a brief explanation in the text when it first appeared.

Section 2

Point 7: The information provided is quite detailed. However, some parts can be confusing, especially for readers unfamiliar with the topic. For example, it might be helpful to define some of the less familiar terms such as "ROI", "VJ method", "CSK method", etc. Also, consider simplifying complex sentences to increase readability.

Response 7: We are sorry for the confusion. We have added a brief explanation in the text when a term first appeared.

Point 8: The explanation of concepts is generally good, but it could be beneficial to offer more context or background for some of the cited studies. For instance, instead of just mentioning the VJ method or CSK method, consider explaining briefly what these methods entail.

Response 8: Thanks for the comments. As mentioned in the previous point, We have added a brief explanation in the text when a term first appeared.

Point 9: In order to emphasize the importance of the subject, please include in the manuscript more information about usage of rPPG in medical domain:

10. https://doi.org/10.1145/3544793.3563406
11. https://doi.org/10.1016/j.patcog.2023.109421

Response 9: Thank you very much for the suggestions. We have adjusted the content of the introduction to include recent literature about the benefits and applications of RPPG (especially in the medical field) (line 54-67 in new manuscript). We also added evidence about using PPG to improve RPPG signals and described our motivation for proposing the algorithm in the context of modern health monitoring (line 68-85 in new manuscript).

Section 3

Point 10: Provide the units of all the values used in each equation.

Response 10: Thanks for your advice. The equations refer to reference 12 (Wang, W.; den Brinker, A.C.; Stuijk, S.; de Haan, G. Algorithmic Principles of Remote PPG. IEEE Transactions on Biomedical Engineering 2017, 64, 1479–1491). Similar to their work, the equations in this paper use scalar variables and omit units.

Point 11: Align the captions of figures along with the journal requirements. Please, check it here and elsewhere in the manuscript.

Response 11: We have adjusted the figures according to your suggestions.

Point 12: Consider merging figures 4,5 and 6 into one figure. At the moment their size is enormous compared to the amount of additional information they bring to the article.

Response 12: Thanks for the comments. We have combined them into one figure (Figure 3 in new manuscript).

Point 13: The section has some minor grammatical errors (e.g., in line 209, "at a fixed sampling rate, 60 fps." is not clear. Do you mean "at a fixed sampling rate of 60 fps."?) and some awkward phrasings (e.g., in line 231, "A stable and controlled light source such as a fill light is commonly used in research as it has little effect on the results." might be better rephrased to "Fill lights, which are stable and controlled light sources, are commonly used in research because they minimally affect the results.")

Response 13: We are very sorry for our incorrect writing. The manuscript has been thoroughly revised and edited, and we hope it can meet the journal’s standard.

Point 14: Study's generalizability is severely limited due to the homogeneity of participants. It only includes participants of a specific age range and skin tone. Diversifying this could improve the relevance of research and allow for more meaningful interpretations.

Response 14: Thank you for the comments. We are also aware of the limitations of the participants. We describe the shortcomings of the study in the discussion section and would like to increase the diversity of subjects in future studies.

Point 15: Section 3.2.5. includes necessary details, such as the key comparison metric and the statistical methods used. However, it might be helpful to specify more clearly the goal of the statistical analysis and why the p-value of 0.05 was chosen as the significance threshold.

Response 15: Thanks for the reminding. We have added some content about ANOVA to explain the meaning of significance in this study. We used 0.05 as the significance threshold because it is a widely accepted choice in many studies.

Section 4

Point 16: Figures 9-12 are not readable at all, please, adjust size and font.

Response 16: Thanks for your careful check. We have adjusted the content of the figures. Only one subject is included in the paper, and the data of the other three subjects are included in the Appendix.

Section 6

Point 17: Please, include a statement regarding future research directions, echoing the points raised in the Discussion section. It might be beneficial to remind the reader of the ongoing and future work that builds upon this study.

Response 17: Thank you for the suggestion. We have fleshed out the conclusions, including a brief background (and related recent work), the content and future directions of this study, and responses to the results and discussion.

Overall comments

Point 18: Unify the figures in terms of size and style. It will improve visual quality. The manuscript can’t go further unless it is done.

Response 18: We really appreciate your comments and they do help a lot. We adjusted the size of the figures and corrected the grammatical errors in the article, hoping to improve readability. We also enriched the introduction and conclusion, including describing the recent application and advantages of RPPG, and supplementing the background and motivation of the research.

Round 2

Reviewer 1 Report

all comments have been addressed 

Reviewer 4 Report

this updated version of the manuscript is fine to be accepted

this updated version of the manuscript is fine to be accepted

Reviewer 5 Report

 

All major comments were adequately addressed and the Authors have done an admirable job of improving the quality of the manuscript. Therefore, it can be accepted without any structural modification.