Automated Building Heritage Assessment Using Street-Level Imagery

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dabrock, Kristina, Johansson, Tim, Donarelli, Anna, Mangold, Mikael, Pflugradt, Noah, Weinand, Jann Michael, Linßen, Jochen
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915683053338624
author Dabrock, Kristina
Johansson, Tim
Donarelli, Anna
Mangold, Mikael
Pflugradt, Noah
Weinand, Jann Michael
Linßen, Jochen
author_facet Dabrock, Kristina
Johansson, Tim
Donarelli, Anna
Mangold, Mikael
Pflugradt, Noah
Weinand, Jann Michael
Linßen, Jochen
contents Registration of heritage values in buildings is important to safeguard heritage values that can be lost in renovation and energy efficiency projects. However, registering heritage values is a cumbersome process. Novel artificial intelligence tools may improve efficiency in identifying heritage values in buildings compared to costly and time-consuming traditional inventories. In this study, OpenAI's large language model GPT was used to detect various aspects of cultural heritage value in facade images. Using GPT derived data and building register data, machine learning models were trained to classify multi-family and non-residential buildings in Stockholm, Sweden. Validation against a heritage expert-created inventory shows a macro F1-score of 0.71 using a combination of register data and features retrieved from GPT, and a score of 0.60 using only GPT-derived data. The methods presented can contribute to higher-quality datasets and support decision making.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11486
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Automated Building Heritage Assessment Using Street-Level Imagery
Dabrock, Kristina
Johansson, Tim
Donarelli, Anna
Mangold, Mikael
Pflugradt, Noah
Weinand, Jann Michael
Linßen, Jochen
Computer Vision and Pattern Recognition
Registration of heritage values in buildings is important to safeguard heritage values that can be lost in renovation and energy efficiency projects. However, registering heritage values is a cumbersome process. Novel artificial intelligence tools may improve efficiency in identifying heritage values in buildings compared to costly and time-consuming traditional inventories. In this study, OpenAI's large language model GPT was used to detect various aspects of cultural heritage value in facade images. Using GPT derived data and building register data, machine learning models were trained to classify multi-family and non-residential buildings in Stockholm, Sweden. Validation against a heritage expert-created inventory shows a macro F1-score of 0.71 using a combination of register data and features retrieved from GPT, and a score of 0.60 using only GPT-derived data. The methods presented can contribute to higher-quality datasets and support decision making.
title Automated Building Heritage Assessment Using Street-Level Imagery
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.11486