Understanding subjective urban experiences is essential for designing cities that enhance well-being. Urban design should account for the psychological effects of environments on individuals, as these significantly shape perceptions and behaviors. However, a major challenge is the limited availability of urban perception data. Recent studies have leveraged large, crowdsourced datasets like Place Pulse 2.0 (PP2) to inform machine learning (ML) models for urban perception prediction, but the accuracy and reliability of outcomes remain underexplored. There is a critical need to evaluate whether these datasets truly capture human perceptions. This study investigates the role of urban street images in understanding environmental perceptions, using the PP2 dataset and ML techniques. It explores various ML pipelines, employing TPot AutoML for model selection and 5-fold cross-validation to prevent overfitting. The goal is to identify the most efficient model that strengthens the link between automated predictions and human perception. The study also applies SHAP (SHapley Additive exPlanations) to interpret model outputs, revealing feature importance and interactions. This improves transparency and ensures ML-generated insights are actionable for urban planning. By rigorously testing ML pipelines, this research enhances predictive accuracy and contributes to the development of reliable urban design tools. The findings highlight ML’s potential in processing large-scale perception data, uncovering hidden patterns, and informing people-centered urban planning. However, further validation against real-world surveys is necessary to ensure robustness and generalizability in assessing urban perceptions.
The Use of the Imagematching Software and Other AI Tools for the Automatic Recognition of Similar Images: Some Theoretical Considerations
The article discusses the new digital research methodologies to investigate the concept of image similarity, taking as a case study the Lyon16ci database project. Developed in collaboration with the Visual Geometry Group at the University of Oxford, this project investigates how AI-driven image recognition can enhance scholarly analysis of visual material in the humanities. The focus is on the use of VISE software, designed to automatically retrieve visually similar images based on geometric and compositional features. The article provides a critical evaluation of the strengths and limitations of this tool in the context of art historical and visual culture research. It discusses how VISE facilitates new interpretative approaches by uncovering visual relationships and how it can effectively enhance traditional comparative methods. The author provides an overview of the advantages and constraints of using the VISE AI software to automatically retrieve similar images, presenting some of the theoretical considerations and the research possibilities provided by image recognition tools. The Lyon16ci case offers insights into the broader potential of machine vision in redefining the scope and scale of such image-based humanities research.
Assessing In-Motion Urban Visual Perception: Analyzing Urban Features, Design Qualities, and People’s Perception
The paper proposes a machine learning and computer vision methodology for evaluating urban visual perception during pedestrian movement. The research combines semantic image segmentation, Place Pulse 2.0 perception datasets, Google Street View imagery, and supervised machine learning models to analyze how urban physical features and spatial design qualities influence people’s perception while walking. The workflow integrates PSPNet semantic segmentation, SVM-based perception prediction, and Pearson correlation analysis to investigate relationships between urban morphology, perceived openness, imageability, enclosure, complexity, and subjective perceptions such as beauty, safety, liveliness, boredom, and depression. The study highlights how directional and panoramic visual fields produce different perceptual outcomes and demonstrates the role of AI-assisted urban analytics in understanding pedestrian experience and informing human-centered urban design.
AR for the Knowledge and Fruition of Street Art Works
The paper presents a methodology for the documentation, preservation, and enhancement of street art through augmented reality technologies. The approach combines photogrammetric survey and image-based tracking techniques to generate high-resolution digital representations of artworks, which are then integrated into an AR application developed with Unity and Vuforia Engine. The system allows users to access contextual information, interpretative content, and multimedia data directly overlaid onto the physical artworks, enhancing their understanding and accessibility. The research is tested on selected street art works in Naples, demonstrating how AR can support both the conservation of ephemeral urban art and the creation of interactive cultural itineraries, while also highlighting the potential for scalable, georeferenced applications and WebAR developments.
Interactive Heritage Site Mobile Application on Artworks
In this work, we are introducing early technological prototypes of an interactive mobile application aimed at heritage sites and museums. This mobile application will comprise three main components as follows: (1) Image Captioning, (2) Indoor Navigation, and (3) Augmented Reality. To address the Image Captioning component, we aim to build a mobile application for the guidance of heritage site visitors with a caption with relevant information about each artwork. This application can be helpful for visually impaired visitors, and it can provide some extra informa-tion about the museum’s objects. We will build this project based on computer vision and AI techniques. The main methodology of this work is according to the Image Captioning algorithms which by giving any input image, the AI model can provide a relevant caption to describe the input image through the textual and voice outputs. For the Indoor Heritage Site Navigation, we propose a mobile application that will be connected to an indoor positioning system which will be able to locate each piece of artwork of the museum and indicate directions to the museum visitors. Furthermore, this application would be beneficial for visually impaired people to follow prop-erly all the artwork of the museum and find their locations. The Augmented Reality technology will provide the interactive aspect of an application for the heritage site visitors to obtain extra information about the artworks. The goal is to show animated visualizations of the artwork and/or to show detailed information about each part of the artwork by pointing out each area and adding some more textual or visual data around that. The proposed application will be a holistic interactive application which will be comprised of all the three above-mentioned packages in one. The key characteristic of this work stems from building a lightweight mobile application to include three platforms together.
Between Image and Text: Automatic Image Processing for Character Recognition in Historical Inscriptions
The research addresses the challenges in Optical Character Recognition (OCR) systems when applied to ancient inscriptions and graffiti. These artifacts, serving celebratory or commemorative purposes, often present legibility issues due to erosion and gaps in the text. Our study proposes an automated image processing pipeline supported by 3D data from photogrammetric surveys. The processing phase involves manipulating image parameters and utilizing spatial coordinates and writing system information. The goal is to enhance legibility by extracting images with neutral backgrounds and highlighted characters, resembling printed texts. This processed data aims to improve the performance of pre-trained Artificial Intelligence (AI) models dedicated to OCR. Ultimately, the research seeks to provide a compar-ative study between unprocessed and processed images, validating the significance of the pre-processing phase in enhancing text recognition systems. The proposed automated workflow aims to contribute to the field of computer vision, specifically in the context of preserving and interpreting historical inscriptions.
Real-Time Identification of Artifacts:Synthetic Data for AI Model
The collections represent the constitutive element and the raison d’être of each museum. Their man-agement, care and dissemination are therefore a task of primary importance for every museum. Applying new Artificial Intelligence technologies in this area could lead to new initiatives. However, the development of certain tools requires structured and labeled datasets for the training phases which are not always easily available. The proposed contribution is within the domain of the construction of specific datasets with low budget tools and explores the results of a first step in this direction by testing algorithms for the recognition and labeling of heritage objects. The developed workflow is part of a first prototype that could be used both in heritage dissemination or gamification applications, and for use in heritage research tools.
Supervised Classification Approach for the Estimation of Degradation
The study presents an innovative approach to classify geomaterials using supervised classification methods from orthophotos derived from UAV (Unmanned Aerial Vehicle) and photogrammetric processing. The case study examined is the Ponte Rotto, dating back to 20 BC, which in antiquity allowed the Appian Way to cross the Calore River – between the provinces of Avellino and Beneven-to – to continue towards the port of Brindisi. In previous studies, experts on geomaterial diagnosis estimated – from aerophotogrammetric orthophotos generated for both bridge elevations – the geo-materials and quantities used for the construction of the monument and an overview of the state of conservation of the monument studied. Orthophotos of facades were imported into CAD software and used as the basis for – according to a manual process – the mapping of the materials. The work presents the results according to automatic Machine Learning clustering from the same orthophotos to identify geomaterials.
Media Convergence and Museum Education in the EMODEM Project
The interpretation of the museums heritage as an active social element is the basis of the most cur-rent cultural institutions projects that envisage forms of documentation, use and dissemination of the cultural heritage increasingly dialogic (the museum community) and dynamics (mixing conversational, experiential, and participative modes).This scenario is the broad framework of the project conceived to design the EMODEM app based on the convergence of face detection, eye tracking and AR to interface the virtual and the physical space and make the museum experience more visitor-centered, interactive and personalized.This article integrates the EMODEM research already underway and updates the scientific roadmap according to the progress recently achieved in phases 3 and 4 of the project, presenting the technolog-ical innovation that has intervened in the meantime in the project and the advancement of research, currently reached at third field usability testing.
“Divina!” a Contemporary Statuary Installation
In 2021, the year of the 700th anniversary of Dante Alighieri’s death, the ASTRO Laboratory of the Pisa University Department of Civil Engineering have set up, in collaboration with the Tuscany Re-gional Council, the contemporary statuary installation “Divina!” based on the work of the great poet. This installation leads users to ponder, from a technological standpoint, the way in which the means of communication are used and the importance of preserving and conserving the roots of linguistic evolution.