Next Article in Journal
Diffusion of Innovation: Case of Co-Design of Cabins in Mobile Work Machine Industry
Previous Article in Journal
An App that Changes Mentalities about Mobile Learning—The EduPARK Augmented Reality Activity
 
 
Article
Peer-Review Record

Procedural Modeling of Buildings Composed of Arbitrarily-Shaped Floor-Plans: Background, Progress, Contributions and Challenges of a Methodology Oriented to Cultural Heritage

by Telmo Adão 1,2,*, Luís Pádua 2, Pedro Marques 2, Joaquim João Sousa 1,2, Emanuel Peres 1,2 and Luís Magalhães 3
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Submission received: 14 March 2019 / Revised: 21 April 2019 / Accepted: 29 April 2019 / Published: 11 May 2019

Round 1

Reviewer 1 Report

The article focuses on the definition of a methodology capable of producing semi-automatically buildings based on ontologies composed of existing arbitrary forms. The article is well structured and presents a deep state of the art on the different types of manual, semi-automatic and automatic modelling. The article starts and resumes a consolidated background from several past experiments, mentioned in the text, but highlighting the progress made compared to previous publications. It seems a bit poor from an iconographic point of view, since the presence of a greater number of significant images would help in understanding the whole process and the application cases. A second point that is not very evident concerns the distance between drawings or image-based models, which represent the basis in the reconstruction phase, and those reconstructed from the procedural point of view. It would be interesting and appropriate to understand if it is possible to identify an average distance between the existing and the reconstructed model, connecting in a stronger way the whole process of survey and 3D modeling and assigning a level of reliability to the virtual models. Finally, the part of the conclusions should be enriched. In fact, in addition to showing the progress of the published research and repeating a description of the projects already reported in the article, it would be more useful to use the space available to specify and describe the conclusions regarding the progress presented in the article and to foreshadow whether there will be further steps in the future, with what timing, structure and what could finally be a result achievable in a few years in the field of procedural modeling in the archaeological field.

From the syntactical point of view, some small inaccuracies listed below:

Line 18 - perhaps it is better "composed by non-rectangular shapes";

Line 43 - I would eliminate "in addition"

Table 1 - In the table some observations:

- In the “short definition” in correspondence of the range scanning I would say that they are not differentiated from the fact of being sensor-based, since even the image-based systems are based on sensors, unless you use the “active” term;

- in the "conservation adequacy" relative to the range scanning I would introduce or simply cite the colour problem, that range scanning is not generally able to reproduce if not with results very far from reality;

- Finally, in the "possible requirements" related to both manual and procedural modeling, it is essential to emphasize the importance (especially in the first) in understanding the problems of geometric construction and morphology modelled

Line 220 - it seems that there was no "of" between "terms" and "time"

Caption Figure 6, second line - correct word "photogrammetric"


Author Response

Thank very much you for your contribute. Please, find your answers in the attached letter.

Author Response File: Author Response.pdf

Reviewer 2 Report

No further comment

Author Response

Thank very much you for your contribute. Please, find your answers in the attached letter.

Author Response File: Author Response.pdf

Reviewer 3 Report

Unfortunately my previous review was not taken on board in this resubmission.

The accompanying letter states "novelty may not be the strength of this paper.". This seems to be true, however it is not acceptable to claim contributions which are decades old as yours:

"• incorporation rules supporting straight skeleton-based roofs, whose delineating polygons might vary between convex and concave shapes;"

As stated in the previous review, the  "Automatically generating large urban environments based on the footprint data of buildings", and CityEngine/CGA have prior art on this. It is quite serious that this claim has not been removed given the resubmission.


The following sentence claims that you created procedural modelling in 2013. Please correct it:

"Procedural modelling methodology [1] work started to be delineated in [13,14] and passed 93 through a few enhancements [15,16] until more recent versions [1,17]. "


The requested images for each step section 3 were not provided. It is still very hard to understand what goes on here. How does L-system string rewriting work to create buildings? Mueller uses L-systems to create streets, and split shape grammars for buildings. Given you spend 8 pages surveying and introducing the problem, perhaps a few pages should be spent explaining (or at least demonstrating) this L-system here, rather than in the cited thesis. Ideally then a comparison to how L-systems + treemaps compare to CGA-Shape's split shape grammar would be presented.

The formula for the area of a polygon is still not required. As requested in the previous review, please remove it. Thank you for defining $i$, somehow you have introduce the newly undefined value $P_{it}$. Still, if you remove the entire section we will have at least solved this problem.


Given the reviewers comments on their sublime quality of english, I will restate my previous comment on the quality of English, and provide examples. I wrote: "I found the quality of English hard going. In general a good proofread **is required**. More specifically, there are too many poorly used idioms - I would suggest writing in shorter, simpler sentences. In particular including simple diagrams for each modeling step would be appreciated (there were too many undefined terms in Figure 2 to be able to understand it)."


This sentence from the second paragraph of the abstract is a good example of over-use of idioms that is hard to follow if english isn't your first language:


"Regarding both image-based and range scanning approaches, besides of usually requiring trained human-resources to prepare field operations and manipulate expensive equipments (e.g. 3D scanners) and advanced software tools (e.g. photogrammetric applications), they are more suitable for digital preservation of well-conserved structures."

Consider a simpler alternate construction (removing the incorrectly used "regarding" and "besides of usually requiring"(?!) idioms), and fixing the plural on "equipment":


"Image-based and range scanning approaches are more suitable for digital preservation of well-conserved structures. However, they both require skilled labour, expensive equipment and specialist tools."


Other statements which are too idomatic or casual (maybe suitable for spoken english, not technical writing) include:

"When virtually addressed, "

"still scarcely addressed"

"In what regards"

"When it comes to"

"Focusing on procedural modelling, during the last decades, the scientific community of this field has been…"

"Comparatively to other modelling approaches,"

"On the one hand, image-based approaches - resorting to, for example, pinhole model or structure from motion (SfM) techniques recurrently applied in photogrammetric processes -  enable the estimation of 3D positions out of 2D digital imagery composed of single or multiple views, which can be obtained either from archives or surveying processes. While data scarcity associated 49 to the former leads " here "in the former" seems to refer to pinhole models?!


Author Response

Thank very much you for your contribute. Please, find the answers to your comments/suggestions in the attached letter.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

No comments, considering the changes applied

Author Response

Thank you very much!

Reviewer 3 Report


from cover letter:


> Indeed, we stated sometimes throughout the paper that we do not claim historical/architectural faithfulness with the proposed methodology.


I agree with reviewer 1: your emphasis on "reconstruction" in the introduction (table 1 etc…) and literature review positions your paper as a reconstruction technique. No evaluation of reconstructions take place. Perhaps rephrasing or shortening this material would help.

From a previous review:

>  In general a good proofread **is required**.
This is still true - the spelling grammar still needs work.


For example "progresses" in the contributions should be "progress"
The correction in my previous review "In what regards" is still present on line 646.

I requested you fix up this sentence (69): "Focusing on procedural modelling, during the last decades, the scientific community of this field has been…" it was corrected to "During the last decades, procedural modelling scientific community has been proposing different…". Please now fixed this to "During the last decades, the procedural modelling scientific community has proposed different..."

Thank you for including Figure 2 to Section 3.3, I now understand what you are trying to convey. To that end, please cite Vanegas et al. - you present an axis-aligned version of their OBB subdivision (Figure six in their paper):


Vanegas, C.A., Kelly, T., Weber, B., Halatsch, J., Aliaga, D.G. and Müller, P., 2012, May. Procedural generation of parcels in urban modeling. In Computer graphics forum (Vol. 31, No. 2pt3, pp. 681-690). Oxford, UK: Blackwell Publishing Ltd.


and mention and compare this technique to CGA's split rules [66].

As in my first review, I requested images for each Step in Section 3. You decided not to provide them.

>The requested images for each step section 3 were not provided.

Because I have to state this explicitly: please add a similar figure to Step 2. The paper does not give an algorithm for the "preliminary generation of the floor-plan", and how it is different to Step 3. Where do the weights come from, what do they mean - it is impossible to read 3.2 without having first read 3.3. Perhaps a figure giving an example and counter-example (of bad weights) will help here too?

As I stated in the previous two reviews:

> The formula for the area of a polygon is still not required.

and yet here is is on line 420. It is high-school geometry and not required in a research journal. It adds nothing to your exposition.

Equations (1) and (2) now detail the "basic split rules" from [66]. Please note this similarity the manuscript, and explain how your system is different.


Generally the technique given in 3.2 is crude - why are we iterating to find such a solution? isn't there a geometric computation we can perform from the desired weights? Again additional figures are required for exposition.


Please also add a figure for 3.4.



As hinted previously, please include the following citation when discussing skeleton roofs:


Laycock, R.G. and Day, A.M., 2003, June. Automatically generating large urban environments based on the footprint data of buildings. In Proceedings of the eighth ACM symposium on Solid modeling and applications (pp. 346-351). ACM.


Author Response

Dear reviewer,


Please, the answers to your comments follows attached.


Thank you!


Best regards,

Telmo Adão

Author Response File: Author Response.pdf

Round 3

Reviewer 3 Report

Thank you for the updated figures and corrections.

>  We did not perceive this hint in the previous revision round. Reference was added (L-117, L-701).

My first review of this paper included the following paragraph: "Skeleton based roofs have been covered in the literature extensively ("Automatically generating large urban environments based on the footprint data of buildings")..."



Back to TopTop