Optimizing Face Recognition Inference with a Collaborative Edge–Cloud Network

Round 1

Reviewer 1 Report

11      The subject matter is interesting but the authors recall well-known tools and protocols.

22.   What is the novelty and originality of this work?

3 3.  In the context of biometrics, and more precisely of facial recognition and edge-cloud network use, the authors should analyze and cite the following key references:

 

Ear Recognition Based on Deep Unsupervised Active Learning IEEE Sensors Journal, vol. 21, no. 18, pp. 20704–20713, 15 Sept.15, 2021, doi: 10.1109/JSEN.2021.3100151

Past, present, and future of face recognition: A review, Electronics, vol. 9, no. 8, 1188, 2020, doi: 10.3390/electronics9081188.

 

3D Face Recognition Using Multiview Keypoint Matching, 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance, 2009, pp. 290-295, doi: 10.1109/AVSS.2009.11.

 

Cloud-Vision: Real-time face recognition using a mobile-cloudlet-cloud acceleration architecture, 2012 IEEE Symposium on Computers and Communications (ISCC), 2012, pp. 000059-000066, doi: 10.1109/ISCC.2012.6249269.

 

Deep Unified Model For Face Recognition Based on Convolution Neural Network and Edge Computing, in IEEE Access, vol. 7, pp. 72622-72633, 2019, doi: 10.1109/ACCESS.2019.2918275

4. This work needs to be more thorough and requires quantitative measures and other metrics.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

See attached file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

In this work, authors have proposed a real -time face recognition inference application which is deployed, using 

the MTCNN detector and python’s face recognition library. Further, the performance of the application was evaluated on an edge device, a cloud device, and the proposed 378

edge-cloud system. The work is intereseted however need substaintial revisions before publication as given below.

1. What is novelity of the work.Please underscore the scientific value added/contributions of your paper in your abstract and introduction and address your debate shortly in the abstract.

2. A good article should include, (1)originality, new perspectives or insights; (2) international interest; and (3) relevance for governance, policy or practical perspectives relevant to the focus of this manuscript. Please emphasis the originality of your work.

3. How edge infrasturure is designe. Provide details.

4. What is the accuracy of face detection method? why it is not discuused in the manuscript. mention it.

5. Python 2.7 or Python 3.3+ can be used for the face recognition library dependency. But need to discuss algorithm utilized in library.

6. Can you implement edge comuting at algorithm level.

7. which dataset is utilized for evaluation.

8. Literature review need to be enhanced.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors took into account our recommendations and answered our questions. The manuscript has been considerably improved and deserves to be published.

Reviewer 2 Report

All of my comments were properly addressed. Well done!

Reviewer 3 Report

Authors have addressed all the major concerns/ revisions in the revised manuscript. Manuscript is acceptable in its present form.