Analytical Evaluation of Midjourney Architectural Virtual Lab: Defining Major Current Limits in AI-Generated Representations of Islamic Architectural Heritage

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I have reviewed this article about the generation of Architectural Images of Islamic Heritage by means of AI.

I consider that the manuscript is interesting in its scope and objectives as it intends to spread the rich Islamic Architectural Heritage and this is remarkable since we have considerably few efforts in that direction.

I do not find relevant problems with English language

The article is however, not sufficiently developed from the technical point of view since the suggestive images presented are not technically sustained and therefore are not buildable nor intended to be built. In such sense, perhaps this work is more appropriate for a Journal on Heritage or focused on Computer Design and Imaging.

There is in fact an insufficiency detected since the article just tends to outline that the Midjourney software is not capable of accurately rendering existing and well-known monuments of Islamic architecture of non-descript provenance.

On the other hand, the core and fundaments of Islamic Heritage architecture are neither explained nor defined in the article in its present form.

We are not informed of the plans and sections and main features which define or constitute such an important architecture.  Only marginal details like a minaret or a calligraphy which, being invented do hardly offer any meaning, contrary to the intentions of the original builders and promoters.

Drawings and graphs are reduced to impressive photographs but they are not sufficiently analytical.

The examples presented are limited and not documented as of their provenance, location and date of completion for instance. Age is not a characteristic that the reader can deduct from a tiny picture. The purpose of this research remains unclear, although I recognise that the work effected is noticeable.

The references seem adequate but could be expanded.

I acknowledge that the authors have worked thoroughly on the matter albeit their article is not really focused on building sciences but more on representational and narrative issues about ancient architecture, with which the outcome is slightly insufficient.

The conclusions of the manuscript are missing the point quite a bit in this sense.

Summary of evaluation: The article is not logically developed from the academic point of view of building sciences for lack of objectivity and technical insights and its purpose is undefined. It should be in my humble opinion re-considered for publication after major corrections.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The revised manuscript can be published in “Buildings”. The research conducted analytical experimentations with an architectural virtual lab, focusing on defining limitations in AI-generated representations of Islamic architectural heritage. The research topic is interesting; however, there are some issues that need to be revised before publication. Therefore, my decision is an acceptance with minor revision.

 

Here are my comments on improving the manuscript:

- Please consider moving the research question in the abstract to the introduction section. If the authors want to keep it, please consider changing the form of this question sentence to a statement sentence format in the abstract section to emphasize why the authors studied this research.

- Moreover, to clarify the introduction, please consider stating the research questions and -objectives.

- Please consider separating the literature review into subsections, including Islamic Architectural Heritage, AI application (in general), and the need to apply AI for Islamic Architectural Heritage.

- Please provide references to support four main thematic groups: limits of the prompt; limits of fame, limits of regionality and historical styles, limits of architectural and urban elements and details.

- Future work is good; however, the conclusion section is too long due to the addition of discussion content. Please consider moving it to a discussion section.

 

- Some references are too old (over ten years). Please replace the new ISI articles to improve this section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

-- What was the aim of this study and what conclusion was reached?

- What is your difference from the studies in the literature?

- How will it contribute to the academic community and studies?

- “It is worth mentioning that Midjourney does not refer to any particular original images used to generate its AI images but only gives the outcome of the prompt in the form of a set of four images. The user can pick one to make a second round of iterations. “In the case of this research, the second iteration did not give dramatically different results but minor variations on the picked images.” You say that your sentences give different results. Which image should we use as reference for the accuracy and sensitivity of the work?

- What is your sensitivity to work on the subject?

- “Consequently, entering the same prompt, even at the same time, by two different users might not necessarily give similar, let alone identical, results.” You mention that the results are variable. Where is the science here?

- This article is a study that reveals what Midjourney can do rather than academic depth.

 

- Working sensitivities, measurements etc. academic results should be included.

 

Deficiencies must be corrected. It can be published after these deficiencies are completed. MAJOR Revision

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I believe that the article has improved significantly and that the authors have followed my suggestions. Still, I miss more architectural drawings especially in the form of plans and sections.

Reviewer 3 Report

Comments and Suggestions for Authors

- The author made the necessary corrections in his manuscript. It is appropriate to publish.