In recent years, the automatic segmentation and classification of digital survey data has been experimented with in built heritage studies. Despite the encouraging progress in the use of Machine and Deep Learning techniques, the semantic segmentation of point clouds is more complex, especially for the historic environment for which, due to the heterogeneity of shapes, it is more difficult to recognize homogeneous regions with similar properties. Given the need to process a large volume of already annotated data for the training and recognition of new scenes, the type and quality of the initial data play a fundamental role in the classification process, as they influence the subdivision into predefined categories that are not always consistent with a decomposition into architectural elements and sub-elements shared by the scientific community. This is an interpretative problem that already emerges from traditional manual labelling, which, being highly subjective, reduces the reproducibility of the results. The paper focuses on understanding to what extent the recognition of homogeneous regions is influenced by factors such as: manual labelling carried out by annotators with different specializations; density value of the point clouds; and type of data acquired depending on the acquisition sensor. These evaluations were conducted by employing the Random Forest algorithm on specific pre-processed point cloud datasets, with reference to the typology of the Franciscan cloister, in order to make the recognition flows more controlled and less ambiguous, aiming at an advancement towards more efficient modelling and management of the existing architectural heritage.
Augmented Reality Application for BIM Maintenance Feedback via Streaming Platforms
The paper proposes a BIM-based maintenance workflow integrating augmented reality, panoramic image tagging, visual programming, and data streaming platforms for facility management. The methodology combines 360° spherical photography, BIM models, Dynamo scripting, Python automation, Speckle streaming, Grasshopper, and Fologram AR interfaces to create a real-time feedback system for building maintenance. Users can report anomalies directly within panoramic virtual tours, while maintenance operators interact with the BIM model through augmented reality interfaces connected to the building database. The workflow enables automated synchronization between physical spaces and digital twins, facilitating damage detection, metadata updating, and collaborative maintenance operations through accessible mobile and AR devices.
Classification and Recognition Approaches for the BIM Modeling of Architectural Elements
This paper investigates automated approaches for the recognition, classification, and parametric modeling of architectural elements within BIM environments. The research focuses on the integration of image-based and point cloud data with algorithmic and AI-driven methods to support the transition from manual Scan-to-BIM workflows to automated digital reconstruction processes.
The proposed methodology combines visual programming, scripting techniques, and machine learning algorithms to segment, classify, and reconstruct architectural components such as columns, wood trusses, and walls. The workflow includes data acquisition, algorithm definition, training and testing phases, and the generation of parametric BIM objects through API-based implementation. Case studies demonstrate how geometric rules derived from architectural treatises and historiographic knowledge can be encoded into parametric systems, enabling both reconstruction and validation of architectural elements. Results highlight the potential of automated classification and recognition techniques to improve efficiency and reproducibility in BIM modeling, while emphasizing the ongoing need for interpretation and refinement due to the complexity and variability of historical architecture.