Implementation of Automated Baby Monitoring: CCBeBe
Round 1
Reviewer 1 Report
The article is on a very interesting topic and it is obvious that today's advances in machine learning must find a way towards practical applications such as the one described in this paper.
The paper is not very well structured/written. Here are some comments/questions:
Where exactly are the deep learning and back end servers located? Are they separate servers? This could be regarded as hardware dependent.
How do you transfer securely sensitive baby image data between servers?
Screenshots as data visualization (figures 9, 12, 13) is not acceptable.
A weighted average of F1 score at 0.63 is not very good for cry detection, please compare to related work.
Considering the real-time performance, please provide some delay measures. E.g. time between the baby moving and that actual motion showing up on screen of parent’s phone.
Author Response
Please refer to the attached file.
Author Response File: Author Response.pdf
Reviewer 2 Report
The article proposes an automated baby monitoring system CCBeBe which will monitor infants’ lying posture and crying and provides to the parent’s video streaming and voice transmission. The features used are OpenPose, EfficientNet, WebRTC, and Facial Expression Recognition. The system is also integrated within an Android application.
Positive points:
- The references are recent;
- The article is well structured and the concepts are well explained;
- The practical side of the article represents an amazing contribution to the scientific community;
- The authors are encouraged to continue their work.
Observations:
Page 9: The figure 16 title is not under the image and it is on the next page
Page 22: The Figure 22a title is not under the image and it is on the next page
Recommendations:
- I recommend that the article should be strongly accepted after the two corrections of the figures.
Author Response
Please refer to the attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
Please, have a paper reviewed by a native English speaker. It will greatly improve the quality of the work. The syntax problems are few but irritating.
Eliminate the claim below. The claim is unsupported in the paper and unjustified regarding the medical research and it misinterprets the function of the proposed system.
Sudden Infant Death Syndrome(SIDS) is a major cause of death of infants worldwide. To 11 prevent unexpected death, we propose an automated baby monitoring system CCBeBe which 12 monitors infants’ lying posture and crying and provides parents-to-baby video streaming and voice 13 transmission.
The paper is about a software system and any references to the medical benefits it may offer are incorrect and false. The only claim that may be formulated is that the system may potentially be applied to monitoring of infants. Any suggestions that it may impact SIDS are serious misinterpretations.
Minor comments in text.
Comments for author File: Comments.pdf
Author Response
Please refer to the attached file.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
All comments have been satisfied