AI- and AR-Assisted 3D Reactivation of Characters in Paintings
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis article presents the findings of research that aims to preserve and interact with 2D painting characters in 3D using Generation in AI and Interaction in AR. Specifically, it presents a case study of painting "New Year's Market in a Time of Peace". The painting was selected from among the artworks on the open-access digital collections of traditional Chinese painting archived by the National Palace Museum, which constitute the scope of the research. The study is significant because it incorporates the combination of AI and AR into a process of reactivate a new preservation paradigm of cultural heritage. Consequently, the text presents an insightful study on technologies that can be utilized to facilitate the reconnection of the past and the present.
In the section entitled "Materials and Methods", the development of the study could be made more detailed. In particular, the processes of training and generation in AI (see pages 3-4, lines 114-167). In this sense, the study reported herein would be enriched by this consideration, as it would allow for a broader discussion on the potential and legitimacy of this technology for the preservation of cultural heritage. This debate, however, is not explored in depth in the text, despite its evident relevance. For instance, the composition and quality of the data utilized in AI training could be examined and discussed in detail. Additionally, the text does not clarify whether the study attempted to avoid historical distortions and ensure respect for the traditions and customs of the time period. This could be essential for associating the study with a new reconstructed contemporary identity of culture. These issues, which also relate to ethics in the use of generative AI, could be included in the sections titled "Results" (pages 6-10, lines 178-237) or "Discussion" (pages 10-15, lines 238-365). Furthermore, the conclusion (pages 15, lines 366-379) could be developed further, incorporating a more comprehensive critical approach to the study, while also including its limitations in addition to its contributions.
Author Response
Dear Reviewer:
Thank you very much for this significant reviewing effort. Please check the attached Word file. Hope I have addressed your comments clearly.
Your suggestions are highly appreciated.
Best regards,
Naai-Jung Shih
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe contribution in question stands out for a proposal of undoubted originality, placing itself in a still little explored field such as the interaction between artificial intelligence, augmented reality and traditional painting, with the aim of reactivating intangible cultural heritage through three-dimensional reconstruction and insertion in contemporary urban contexts. It is a creative and bold idea, worthy of attention for its ability to stimulate new interpretative and applicative paths in the field of digital heritage.
However, despite the richness of content and the variety of cultural and technical references, the text presents a series of critical issues that currently hinder its full academic maturation. The theoretical treatment, although present and sometimes suggestive (graffiti, urban art, visual semiotics), is dispersive and lacks a real argumentative direction. There is no common thread that coherently connects the theoretical background to the heart of the presented project. Furthermore, the text does not explicitly articulate either a research question or a methodological framework recognizable according to the canons of empirical or experimental research. The description of AI/AR processes remains confined to a technical register, without being problematized or critically analyzed.
From the point of view of discussion, the work alternates interesting observations and sharp intuitions with more descriptive and sometimes redundant moments. The absence of a solid argumentative structure — capable of leading the reader from theoretical premises to results and finally to well-founded conclusions — makes the reading discontinuous. The results, although extensively illustrated through images and tables, are not expressed according to criteria of verifiability or replicability, and are not accompanied by metrics, benchmarks or qualitative/quantitative validations. The contribution moves in a gray area between design demonstration and academic research, without clearly adhering to either register.
The bibliography is extensive, updated and touches on both technical and cultural aspects. However, the dialogue with the sources often appears illustrative rather than analytical; few references are critically compared, and there is a lack of works that directly address the topic of AI/AR applied to traditional painting for museum or heritage purposes. Even the conclusions, although animated by enthusiasm and good intuitions, are weakly supported by the data presented and lack a self-critical reflection on the limits of the proposed approach.
In summary, the work stands out for its originality and potential, but requires an in-depth review on the structural, theoretical and methodological level in order to aspire to publication in a high-level scientific journal. It is therefore suggested to consider a substantial reformulation, with the invitation to clearly explain the research objectives, to strengthen the theoretical framework, to reorganize the results according to a verifiable method and to streamline the text to improve its readability and argumentative rigor.
With the hope that these observations can contribute to the improvement of the work, a substantial revision is therefore proposed before any possible acceptance.
Author Response
Dear Reviewer:
Thank you very much for this significant reviewing effort. Please check the attached Word file. Hope I have addressed your comments clearly.
Your suggestions are highly appreciated.
Best regards,
Naai-Jung Shih
Author Response File: Author Response.pdf