This article examines how artificial intelligence (AI), especially generative AI, is reshaping architectural design, historic buildinginformation modelling (HBIM), and digital reconstruction in cultural heritage contexts. It clarifies a central distinction: AI outputs may be visually persuasive, but they become architectural or heritage knowledge only when they can be validated against historical evidence, spatial and material constraints, and conservation requirements. The article brings togethertwo bodies of literature: generative AI in architectural representation, early design, spatial exploration, and studio practice; and AI-supported HBIM, scan-to-HBIM, semantic enrichment, digital twins, and evidence-based reconstruction. Methodologically, this study is a thematic narrativeliterature review and conceptual synthesis, not an empirical case study, systematic review, or bibliometric analysis. The literature is analysed through three lenses: design stage, information form, and epistemic claim. The review shows that generative AI is most visible in conceptual design, visualisation, prompt-based exploration, and rapid production of alternatives. By contrast, AI-supported HBIM research is concerned mainly with point-cloud processing, semantic segmentation, scan-to-HBIM automation, ontology-based enrichment, data quality, and heritage-management workflows. This comparison reveals a persistent gap between thespeed of representation and thereliability of information. In heritage contexts, the gap is critical because HBIM models and reconstructions can guide conservation decisions, interpretation, and public memory.The article concludes that AI-supported heritage modelling should be assessed usingevidence, uncertainty, authenticity, provenance, metadata, paradata, and expert validation, rather thanvisual plausibility alone. Its original contribution is to frame evidence-based generative HBIM as an operational validation workflow. In this workflow, AI-generated elements remain provisional until they are linked to sources, marked by uncertainty, checked againstsurvey data, archival evidence, typological and material knowledge, and expert judgement, and integrated into HBIM as traceable information rather than as anautonomous reconstruction.
El estudio tiene como objetivo principal desarrollar una reconstrucción virtual tridimensional (3D) integral de la ciudad prehispánica de Oxtankah, en la bahía de Chetumal. Los fines son de conservación, investigación arqueológica, divulgación científica y mejora de la experiencia del visitante. El proyecto plantea como eje central la integración de la arqueología virtual como herramienta para interpretar, comunicar y gestionar el patrimonio cultural maya. Se priorizan el rigor científico, la transparencia y la verificabilidad. El trabajo adopta un diseño metodológico mixto que combina técnicas arqueológicas tradicionales con tecnologías digitales avanzadas. La recolección de datos incluyó levantamientos topográficos de alta precisión mediante GNSS RTK, estaciones totales, drones y fotogrametría; además, se cuenta con el análisis de informes arqueológicos previos y de modelos digitales de elevación (MDE) con resoluciones ≤1 cm/píxel. El procesamiento y modelado se realizó en gabinete mediante software especializado (AutoCAD, QGIS, Surfer, Agisoft Metashape y Blender 4.02), siguiendo un flujo de trabajo estructurado en seis fases: planimetría, modelado poligonal, mapeado UV, texturización PBR, iluminación y renderizado. Las reconstrucciones se fundamentaron en evidencias arqueológicas, analogías regionales y criterios establecidos por la arqueología virtual. El proyecto generó un modelo 3D detallado de 440 estructuras arqueológicas e históricas dentro del polígono del INAH, incluyendo plazas, conjuntos habitacionales, áreas productivas, paisaje natural y cuerpos de agua. Se incorporaron personajes digitales y escenarios de la vida cotidiana para contextualizar el entorno sociocultural. Los modelos se aplicaron directamente al diseño de 27 paneles informativos, museografía y señalética, optimizando la comprensión espacial y cronológica del sitio. Los resultados demuestran que la reconstrucción virtual constituye una herramienta eficaz para la preservación digital, la generación de hipótesis reconstructivas y la divulgación del conocimiento arqueológico. Aunque las reconstrucciones se reconocen como propuestas hipotéticas sujetas a actualización, su aportación al estudio del paisaje cultural maya y a la gestión del patrimonio es significativa. Todo ello consolida la arqueología virtual como un enfoque metodológico de alto valor interdisciplinar.
This study proposes a methodology that combines traditional two-dimensional (2D) drawing and three-dimensional (3D) modelling methods with generative artificial intelligence (GenAI) for the virtual reconstruction of destroyed buildings. This methodology can reduce production and texturing times without compromising the historical accuracy or photorealistic quality of the final images. As a case study, we propose to reconstruct the crypt of the Royal Monastery of Saint Engracia in Zaragoza, Spain, a space of Early Christian origin dating from the fourth century AD that reached its greatest extent in the 17th century. The complex was destroyed in the 19th century during the French invasion of 1808-1814 and subsequent interventions removed the surviving architectural remains. Consequently, no physical remains are available as a basis for the reconstruction. The methodology is therefore based on the analysis of historical documentation and architectural analogues. An initial floor-plan hypothesis was developed using AutoCAD 2025, followed by the creation of a detailed 3D model in Rhinoceros 8. In the final stage, the model was rendered without materials or textures in D5 Render. The resulting images were then processed using Nano Banana Pro, Gemini's image-generator tool, to generate textures and wall paintings with GenAI. Although the results were satisfactory, GenAI was found to introduce arbitrary artefacts or random interpretations into textures and architectural elements. For this reason, the tool should not operate autonomously, as human supervision is required to filter and validate each action and ensure that automatic generation does not compromise technical and historical accuracy. In conclusion, integrating GenAI into virtual reconstruction not only accelerates the workflow, but can also allow us to overcome the technical obstacles that hinder the virtual reconstruction of lost buildings, provided that its use is guided by scientific criteria and human oversight.
This article presents a work about the coffered ceiling of the Synagogue of El Tránsito (Toledo), an exceptional example of 14th-century Spanish-Muslim carpentry. Its structural model follows the typical geometric standards of the period which were common to both Christian and Muslim traditions in the 14th-century in Spain. This work includes, on the one hand, the analysis of the coffered ceiling geometric design: first, a basic module has been identified, as well as the several patterns that were developed with its variations and combinations; secondly, the whole design has been fit to the global dimensions to generate an idealised whole layout of the structure. Finally, a comparison with the actual geometry of the building has been carried out to analyse the reasons of the differences between them. On the other hand, a study of the coffered ceiling constructive configuration has been developed: the pieces that form the structure have been firstly individualized to study their assembling process later. The research has been possible thanks to a three-dimensional (3D) survey done in the building combining several techniques (laser scanning and photogrammetry). This generated a point cloud which has been the geometric basis for the latter analysis and for the generation of 3D parametric models of the coffered ceiling pieces. The main results are the geometric layout which could be used by carpenters to design the coffered ceiling, as well as the explanation of the wooden pieces and their assembling process. The paper is fully illustrated with two-dimensional (2D) drawings and images of the 3D models, thus helping to reach a better knowledge, conservation and dissemination of these types of historical wooden structures.
Three-dimensional Gaussian Splatting (3DGS) enables the high-fidelity reconstruction and rapid rendering of physical objects and spaces while producing highly portable files, enhancing immersive 3D cultural heritage visualizations. Although existing quality metrics demonstrate technical excellence in encoding accuracy and perceptual quality, stakeholders’ perspectives on this technology and their willingness to adopt it remain largely unexplored. A human-centered evaluation is therefore needed to assess the technology’s practical viability in cultural heritage preservation contexts. Using questionnaires and semi-structured interviews supported by static and interactive visualizations of Puri Klungkung, a Balinese palace in Indonesia, we collected quantitative and qualitative data from eight local experts in history, archaeology, vernacular architecture, and heritage conservation. Our results suggest that experts evaluate 3DGS and photogrammetry differently in both static and interactive visualizations. 3DGS was consistently preferred for close-up views containing complex elements, such as vegetation, water, and architectural details, and was perceived as providing better detail preservation, visual quality, and geometric accuracy. The two techniques received comparable ratings for their representation of materiality and spatial perception. Photogrammetry was preferred for scholarly dissemination because of its greater geometric accuracy and level of detail. By contrast, 3DGS in its current form, appears better suited to public-facing projects, where its visual elements generate more photorealistic renderings, although its geometric accuracy may be lower. Considering the target users’ needs from the outset of the project is therefore important when selecting the most appropriate visualization technique. The thematic analysis yielded recommendations concerning applicability, user familiarity, visual properties, interactivity, and practical implementation, guiding effective visualization of historical Hindu cultural heritage sites in South and Southeast Asia.
