Underwater 3D Reconstruction from Video or Still Imagery: Matisse and 3DMetrics Processing and Exploitation Software
, Maxime Ferrera 1, Javier Escartín 2, Marjolaine Matabos 3, Nuno Gracias 4 and Jan Opderbecke 1Round 1
Reviewer 1 Report
- Matisse & 3DMetrics are 2 years old. Why do you present 2 years old software? For promotion or commercialization?
- Lines 7-8: I wonder about 4 datasets. Is it actually one dataset with 4 parts (sets) of data?
- Font in Fig. 3 is too large.
- You didn't state what is the novelty and the contributions of the paper. Is it in a new dataset?
- The paper could be interesting if rewritten and better presented. IT should be organized to be easy to follow for readers.
- Only 1 reference from 2021-23? You should improve reference list and remove some of self-citations.
- What are the scientific contributions of the paper? It should be clearly stated in the Introduction.
- What are the goals of the paper? If the purpose is to present novel dataset, the paper should be differently organized.
- The paper seems a bit confused. You should try to rectify that by rewriting or better presentation of your work.
Author Response
Dear Reviewer, you will find our answers in the pdf attached.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The content of the article is consistent with the scientific area of the Journal of Marine Science and Engineering.
The subject raised by the authors is current and so far rarely noticed.
The paper has an original and research nature.
The authors are dealing with reconstruct submarine 3D textured scenes from underwater video or still imagery.
They present an open-source software, Matisse, which is suitable for non-expert users and to carry out routinely.
This software facilitates submarine image data exploitation and analyses,
I highly rate the delivered a datasets of different 3D scenes, including both input imagery and navigation, and the associated 3D scenes – the are very good for another authors, which want to compare their results with this authors.
For a better clarification, please edit your paper as follows:
Red mark - There is a cycle or alternative (if)
Edit the flowchart - I recommend using the standard flowchart tags.
I recommend publishing the article after the proposed modifications.
Author Response
Dear Reviewer, you will find our answers in the pdf attached.
Author Response File: Author Response.pdf
Reviewer 3 Report
All abbreviations, even if commonly known in the community, should be explained at first occurence (ROV, AUV - line 17, SFM - line 72, SIFT - line 150).
Please use unique writing of SFM (sfm, SfM, SFM, lines 72, 156, 174).
line 7: ´"four" instead of "4", line 15: delete blancs and ")"
line 159: use small letters for exhaustive, transect, and navigation
Fig. 5: differences between left and right image are hardly visible
A paragraph concerning potential improvements is missing in the perspectives section: can the processing time considerably reduced? How or why not?
A bit more discussion of the possibilities and limitations of the software tool would be convincing.
Author Response
Dear Reviewer, you will find our answers in the pdf attached.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
- You should reduce number of self-citations (13).
- In answer 1.7: "it reposes in state-of-the-art research in computer vision and robotics" - which? It should be explained in details. Are there any novelties in it? The point of scientific paper is to present scientific contributions. Is ti possible that some of the functions are based on some novelty?
- There is a problem in judging your paper. For example, if only new program is presented, how can I grade "Scientific Soundness"? You should make more efforts to provide some reasons to accept the paper.
- Regarding answer 1.1: Does it mean that you can publish paper each time you make some small contribution to the program?
Author Response
Please find our answers in the document attached.
Author Response File: 
Author Response.docx
