Automatic Reconstruction of 3D Models from 2D Drawings: A State-of-the-Art Review
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis article provides a comprehensive literature review on the generation of 3D models from 2D drawings, which plays an important guiding role in subsequent related research. The following issues need further modification and improvement.
1. In line 143-166, is there any other noise that has been removed? Currently, the main type of noise in this paper is external influences, but the drawing itself may cause noise due to issues such as labels and overlapping text.
2. In line 168, is there a relationship between the recognized text and the language?
3. Line 228, is there a relevant standard specification or technology that can achieve automatic layering.
4. In line 278, can the rule-based approaches meet the expression requirements of most components? Generally, standardized components are easy to identify, but how should non-standard component rules be considered?
5. In section 3.3, it is recommended to suggest comparing the advantages and disadvantages of different methods and analyzing the possibility of combining different methods.
6. Suggest increasing research and analysis on text recognition. Simply generating a 3D model from a 2D image requires a lot of semantic information to be obtained from the text.
7. In this paper, it is suggested to further distinguish whether the Cad drawings are in PDF or DWG. Because the technical methods and difficulties of these two types are different.
8. In section 4.4, for generating 3D models from 2D drawings, it is generally necessary to accumulate all data. Suggest to add an analysis of the reasons for the lack of research on other components.
9. Limitation analysis is a crucial part of summarizing the shortcomings of current methods and guiding future research directions. Suggest further supplementation and improvement. For example, differences in different professions, issues of accuracy and efficiency, etc.
10. This article is a review and analysis of existing research, but there is limited literature in the past three years. Suggest adding to ensure the timeliness of this state-of-the-art review.
Comments on the Quality of English LanguageMinor editing of English language required
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper “Automatic Reconstruction of 3D Models from 2D Drawings: A State-of-the-Art Review”, as the title implies, presents a review of existing methods for automatically generating 3D models from 2D architectural drawings. The authors highlight the importance of this process, providing information about its affordability and accessibility compared to other 3D reconstruction techniques.
The authors emphasize the need for a critical analysis of current methods, particularly regarding their coverage and completeness in reconstructing 3D models. This analysis aims to identify research gaps and opportunities for future advancements in the field. This fact deserves prize.
The review concludes by highlighting the limitations of current methods, particularly their inability to generate fully comprehensive and semantically rich 3D models. It emphasizes the need for further research to improve 3D modeling performance, specifically focusing on automating the extraction of building information beyond just structures and facades. The development of a tool capable of accurately identifying all elements within a 2D drawing is identified as a key area for future development.
The structure and design of this article are quite good. The authors have organized the study in a logical and clear manner and their research methods are appropriate. The article reads well, no major issues detected regarding the language. The figures used in the article are effectively designed for communicating the key findings. Overall, the study is well-conducted.
However, we are almost in the middle of the year 2024, yet there are no latest positions in the bibliography:
0 citations from 2023, only 2 from 2022
I recommend adding some of the latest research,
for example:
Zhang, C., Pinquié, R., Polette, A., Carasi, G., De Charnace, H., & Pernot, J. P. (2023). Automatic 3D CAD models reconstruction from 2D orthographic drawings. Computers & Graphics, 114, 179-189.
Pan, Z., Yu, Y., Xiao, F., & Zhang, J. (2023). Recovering building information model from 2D drawings for mechanical, electrical and plumbing systems of ageing buildings. Automation in Construction, 152, 104914.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Comments and Suggestions for AuthorsThis article provides an overview of existing sources of literature on methods for creating 3D models from 2D drawings. Methodology of 3D reconstruction is provided is provided in a general form. Application statistics are given Semantic Richness of the reviewed literature and Geometric coverage of literature publications. This engineering problem is not new, but remains relevant for the use of BIM technology. The methodology is presented in a generally understandable form. The statistics of source processing in various directions are given. The work lacks demonstration of examples of application of methods by instrumental means, which is a certain shortcoming of this study.
Recommendation to the authors. It is necessary to provide several clear examples of the results of using the above methods.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have made revisions based on the last comments. It is recommended to modify and improve the format of the references according to the requirements of this journal. No new opinions on other aspects.
Comments on the Quality of English LanguageMinor editing of English language required
