In recent years, the demand for flexible and sustainable strategies in digitization processes has represented a significant challenge for the heritage documentation research community. In particular, the tasks of parametric modelling and AI-based semantic enrichment operations, necessary but traditionally time-consuming, is extremely onerous from a user-oriented perspective. Many efforts of the research community have been dedicated to enhancing efficiency through automation, and one of the possible solutions is represented by the employment of machine learning strategies. This study introduces an innovative methodology that integrates Visual Programming Language platforms and 3D Python libraries, thereby implementing the Scan-to-BIM approach. Two case studies - characterized by varying scales, resolutions, and accuracies - have been analysed to validate the proposed pipeline, demonstrating its flexibility and scalability across architectural objects and archaeological assets belonging to museum collections. The workflow involves several steps, starting from classified 3D and 2D data segmented using machine learning techniques with the aim of managing semantically enriched reality-based data in BIM/HBIM environment without scarifying accuracy criteria. Results highlight the methodology's efficiency and adaptability in diverse contexts, offering a compelling alternative to labour-intensive Scan-to-BIM processes. Ultimately, this methodology contributes to the automation in cultural heritage digitisation, underlining the need for comprehensive standards and protocols in this dynamic domain.

A SCALABLE APPROACH FOR AUTOMATING SCAN-TO-BIM PROCESSES IN THE HERITAGE FIELD / Avena, Marco; Patrucco, Giacomo; Remondino, Fabio; Spano, Antonia. - In: INTERNATIONAL ARCHIVES OF THE PHOTOGRAMMETRY, REMOTE SENSING AND SPATIAL INFORMATION SCIENCES. - ISSN 2194-9034. - ELETTRONICO. - XLVIII-2/W4-2024:(2024), pp. 25-31. [10.5194/isprs-archives-XLVIII-2-W4-2024-25-2024]

A SCALABLE APPROACH FOR AUTOMATING SCAN-TO-BIM PROCESSES IN THE HERITAGE FIELD

Avena,Marco;Patrucco,Giacomo;Remondino,Fabio;Spano,Antonia
2024

Abstract

In recent years, the demand for flexible and sustainable strategies in digitization processes has represented a significant challenge for the heritage documentation research community. In particular, the tasks of parametric modelling and AI-based semantic enrichment operations, necessary but traditionally time-consuming, is extremely onerous from a user-oriented perspective. Many efforts of the research community have been dedicated to enhancing efficiency through automation, and one of the possible solutions is represented by the employment of machine learning strategies. This study introduces an innovative methodology that integrates Visual Programming Language platforms and 3D Python libraries, thereby implementing the Scan-to-BIM approach. Two case studies - characterized by varying scales, resolutions, and accuracies - have been analysed to validate the proposed pipeline, demonstrating its flexibility and scalability across architectural objects and archaeological assets belonging to museum collections. The workflow involves several steps, starting from classified 3D and 2D data segmented using machine learning techniques with the aim of managing semantically enriched reality-based data in BIM/HBIM environment without scarifying accuracy criteria. Results highlight the methodology's efficiency and adaptability in diverse contexts, offering a compelling alternative to labour-intensive Scan-to-BIM processes. Ultimately, this methodology contributes to the automation in cultural heritage digitisation, underlining the need for comprehensive standards and protocols in this dynamic domain.
File in questo prodotto:
File Dimensione Formato  
isprs-archives-XLVIII-2-W4-2024-25-2024.pdf

accesso aperto

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Creative commons
Dimensione 8.65 MB
Formato Adobe PDF
8.65 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2988661