This document presents the 3nd International Conference on Visual Pattern Extraction and Recognition for Cultural Heritage Understanding (VIPERC 2024), a premier forum for presenting academic and industry papers on big data, data mining and knowledge discovery, machine learning and deep learning, 3D modeling and simulation (structural, thermal, functional and acoustic analysis), software engineering, and, more generally, aspects related to digital transition for visual pattern extraction, analysis, and recognition to preserve cultural heritage.
Cultural heritage encompasses all tangible and intangible elements of historical, archaeological, architectural, and artistic significance. Each piece of cultural heritage narrates the legends and traditions of people, families, communities, and nations around the globe. It serves as a valuable link to society's origins, helping to shape the future by drawing from the past. Cultural heritage is an integral part of daily life, evident everywhere from ancient ruins to contemporary natural landscapes and art. Tangible cultural heritage includes ancient buildings, archaeological sites, monuments, sculptures, paintings, coins, underwater ruins and cities, shipwrecks, manuscripts, photographs, films, and other items of artistic, archaeological, architectural, and historical importance. Intangible cultural heritage encompasses the acoustic heritage of ancient buildings, traditional crafts and festivals, oral traditions and expressions, dialects and sub-dialects, music, culinary traditions, and ways of life. The process of discovering and representing knowledge from cultural heritage primarily involves extracting, recognizing, and modeling visual patterns, which are crucial for analyzing and exploring hidden features, new hypotheses, relationships, trends, and modes within cultural heritage data. A visual pattern refers to any characteristic perceivable by the human senses. Today, the extraction, recognition, and modeling of visual patterns are increasingly achieved through digital transformation, utilizing modeling and automation techniques, intelligent systems, soft computing, and big data analytics in various real-world scenarios.
From all aforementioned, the 3rd International Conference on Visual Pattern Extraction and Recognition for Cultural Heritage Understanding (VIPERC 2024) aimed to be a satellite event of the IEEE 20th International Conference on Automation Science and Engineering (CASE 2024), presenting academic and industry papers on big data, data mining and knowledge discovery, machine learning and deep learning, 3D modeling and simulation (structural, thermal, functional and acoustic analysis), software engineering, and, more generally, aspects related to digital transition for visual pattern extraction, analysis, and recognition to preserve cultural heritage. VIPERC 2024 welcomed contributions from various research areas related to the digitization of cultural heritage, building recovery, conservation, and maintenance, leveraging current technologies and the support of robotics and automation. It also favored the participation of young research scholars who wanted to present their early-stage in-progress academic work.
VIPERC 2024 was part of the research activity of the Italian National Research Project: "MUD-MADE: MUlti-objective optimization of Digitally MAnufactureD earth building components supported by neural networks", PRIN PNRR 2022.
Topics of interest included, but were not limited to:
• Disruptive technologies -Machine learning and data science for cultural heritage multimedia data -Artificial intelligence in the recovery, analysis, and enhancement of cultural heritage -Augmented and virtual reality systems -3D printing -Internet of things in cultural heritage • Diagnostics -Intelligent sensor systems for building monitoring, recovery, conservation, maintenance and art restoration -Discrete geometry techniques for pattern recognition and pattern matching in cultural heritage images -Image processing, texture, shape analysis and natural language processing in historical data
• Acoustics and digital technology -Acoustics in cultural heritage -Soundscape planning and preservation
-3D reconstruction and model processing -3D modelling and simulation (e.g. structural, thermal, functional and acoustic analysis) for cultural heritage -Remote sensing for cultural heritage preservation -Surrogate models (computational models that approximate the behaviour of complex systems or processes)
• Manufacturing and intervention -Robotics and digital manufacturing for built heritage -Digitization of industrial products, projects, prototypes, and artefacts for cultural heritage preservation -Cutting-edge and sustainable digital-supported materials for built heritage -Metamaterials fabricated with specific patterns -Climate changes and intelligent mitigation strategies to preserve cultural heritage sites
In the past, four VIPERC events were organized. Two of them were in the form of international workshop, and two of them were in the form of international conference:
• The 1st Both VIPERC 2019 and VIPERC 2020 were in-site and co-located with the Italian Research Conference on Digital Libraries (IRCDL 2019 and2020). The book of Proceedings of the two events is available in CEUR-WS (Volumes 2320 and 2602).
The attendance rate of VIPERC events in previous years made it possible to offer the next VIPERC events in the form of conference rather than workshop. In particular, VIPERC 2022 was in virtual form and co-located with the 19th International Conference on Artificial Intelligence: Methodology, Systems, Applications (AIMSA 2022). Also, VIPERC 2023 was organized as inpresence satellite conference of the 27th International Conference on Theory and Practice of Digital Libraries (TPDL 2023). The book of Proceedings of the two events is available in CEUR-WS (Volumes 3266 and 3600).
Finally, VIPERC 2024 was organized as in-presence satellite conference of the prestigious IEEE 20th International Conference on Automation Science and Engineering (IEEE CASE 2024), one of the three flagship conferences of the IEEE Robotics & Automation Society, and was held on 1 September 2024 in Bari, Italy.
The scientific relevance of the conference is assured by an international Organizing Committee which includes 8 researchers, six of them are from Italy and two of them are from Croatia, and an international Program Committee which includes 30 researchers from 12 different countries worldwide (Italy, France, Sweden, Greece, Bosnia and Herzegovina, Brazil, South Africa, Finland, Romania, Bulgaria, Serbia, Macedonia).
All members of the Scientific Committee are acknowledged as experts in cultural heritage, knowledge representation, information retrieval, algorithms, pattern recognition, artificial intelligence, and modeling and simulation techniques.
• Alessia Amelio, Senior Researcher, InGeo University "G. d'Annunzio" Chieti-Pescara, Italy, • Drahomira Cupar, Assistant Professor, DIS University of Zadar, Croatia, • Valentino Sangiorgio, Senior Researcher, InGeo University "G. d'Annunzio" Chieti-Pescara, Italy.
Program Chairs:
The VIPERC 2024 Organizing Committee was honored to have invited Professor Stefano Ferilli from the University of Bari Aldo Moro, Italy, to deliver a brief lecture on the conference topics.
Prof. Stefano Ferilli's title of the talk was: "LPG-based Knowledge Graph Representation and Management for Cultural Heritage".
The traditional record-based approach used in Cultural Heritage (CH) has successfully served the need of users, scholars, researchers, and practitioners for many decades, but it has gone as far as it could. The enormous growth in production, types and availability of cultural items, the opening of their use to a wider public (with different background, goals, and perspectives), the advent of digital technologies, and the convergence of many different, traditionally separate disciplines call for new, advanced organization strategies and ways of exploitation, for both the items and the information they carry.
It is necessary to move from a set of predefined fields to a reticular approach, in which each of the different kinds of entities involved in a description has its own record and can be related to the others. In such a new setting the focus of the descriptions can also be broadened, from a fixed set of formal parameters about the items to a larger and more variable set also including information/metadata concerning their physical support, content, context, and even lifecycle. This 'holistic' approach allows us to support the needs and activities of different kinds of stakeholders.
In Artificial Intelligence (AI), the networking of data is the core of knowledge, and coupled with ontologies has been successfully used as a representation for Knowledge Bases (KBs), obtaining so-called Knowledge Graphs (KGs). KGs have been thoroughly studied in the Knowledge Representation (KR) branch of AI, interested in Ontologies and Automated Reasoning. Research in KR developed its own solutions for representing and storing knowledge, that have departed from the mainstream solutions for DBs. However, rooting KG data representation and storage in DB technology would ensure optimization and efficiency in data storage and handling. This requires advanced DB solutions, and some kind of cooperation with the solutions coming from KR, that may boost the effectiveness of data management to support the needs of different kind of users, providing them new possibilities for data exploitation and unprecedented opportunities to carry out their activities.
The GraphBRAIN framework was proposed for this setting, in which Formal Ontologies play the role of data schemas for LPG-based DBs. This enables the combined use of different kinds of AI algorithms to the available knowledge, from Graph Mining to Network Analysis, to Automated Reasoning (possibly combining multiple inference strategies). The proposed framework was implemented in a library and API that can be used by many different domainspecific application. A generic platform was developed for performing the general operations on the knowledge, including standard CRUD ones. The framework and technology were applied to several Cultural Heritage branches including history of computing, digital libraries and archives, linguistic analysis.
The VIPERC 2024 conference organized a ceremony to announce the best paper and best presentation awards.
The best paper was chosen by an Award Committee appointed by the General Chairs. This committee included various researchers from the Program Committee who were not affiliated with the authors of the papers. The Award Committee assessed the scientific quality and innovation of the accepted papers, as well as their evaluations from the review process. Based on this assessment, they ranked the papers and selected the top-ranked paper as the best paper. The best presentation was selected by the General Chairs based on the clarity of exposition, quality of presentation, and the ability to effectively communicate the innovative aspects of the work.
The best paper award of VIPERC 2024 was assigned to Cristina Cantagallo, Giorgia Cianchino, Maria Giovanna Masciotta, Donato Palumbo, Massimiliano Pepe, Leonardo Cangelmi and Enrico Spacone, for their paper by title: "Documentation and rapid assessment of the health status of historic centers by the use of 360-degree videos and G.I.S. ". The motivation for this choice by the Award Committee was the following: "The paper presents an outstanding contribution which successfully intersects different aspects, such as innovative application of low-cost technology, contribution to the cultural heritage preservation, and sustainability and resource efficiency. The aforementioned strengths made this paper a clear choice for the award. The approach is innovative, combined with a strong commitment to sustainability and cultural heritage, and constitutes a significant contribution to the field. "
The best presentation award of VIPERC 2024 was assigned to Francesco Fieni, for presenting his paper by title: "Parametric Algorithm-Driven Robotic Fabrication of Minimal Surfaces using Terrazzo Slabs from Marble Waste".
The VIPERC 2024 conference ensured top-quality content by: (i) collecting high-level contributions from various research areas relevant to the digitization of cultural heritage, including image processing, data mining, artificial intelligence, software engineering, and 3D modeling and simulation; (ii) involving esteemed colleagues from different countries in the fields of cultural heritage, artificial intelligence, and pattern recognition in the Organizing and Program Committee, while also maintaining gender balance; (iii) inviting an internationally renowned speaker who is a leader in the event's topics; and (iv) conducting a fair peer review process based on clarity, topic relevance, and methodology.
VIPERC 2024 took place on September 1, 2024, from 9:00 AM to 1:30 PM (registration, welcome speech, invited talk, Session 1, and Session 2) and from 2:30 PM to 6:00 PM (Session 3, Session 4, Award ceremony, and closing) -Italian time -at the Nicolaus Hotel (room T2, floor -1), Via Cardinale Agostino Ciasca 27, 70124 Bari, Italy. The registration fee for each participant was set at 60 euros.
The event was attended by two General Chairs, Dr. Alessia Amelio and Dr. Valentino Sangiorgio, and three Program Committee members, Dr. Cristina Cantagallo, Dr. Alessandro Ricciutelli, and Dr. Samantha Di Loreto. On September 1, 2024, at 10:00 AM (Italian time), Dr. Alessia Amelio and Dr. Valentino Sangiorgio presented the welcome reception, followed by an invited talk by Prof. Stefano Ferilli from the University of Bari Aldo Moro. Dr. Valentino Sangiorgio chaired Session 1 on "3D Modeling and 3D Printing," Dr. Cristina Cantagallo chaired Session 2 on "Digitization of Diagnostics for Historic Constructions, " Dr. Alessia Amelio chaired Session 3 on "Machine Learning and Pattern Recognition in Cultural Heritage, " and Dr. Alessandro Ricciutelli and Dr. Samantha Di Loreto chaired Session 4 on "The Digital Transition in Architectural Engineering. "
The conference received 24 submissions, reviewed by 25 international research scholars from Italy, Pakistan, and Romania. Each submission was reviewed by at least two scholars from different institutions than the authors. Reviewers were not involved in co-authorship with the paper's authors.
The peer review process was conducted using the EasyChair system, evaluating each paper on clarity, topic relevance, and methodology. Only papers with at least two acceptance scores and no reject scores were accepted. The accepted papers were authored by 48 research scholars from Spain, India, Italy, and South Korea, resulting in 11 full papers included in the Proceedings.
Session 1 featured three presentations on the digital transition for 3D modeling and 3D printing in cultural heritage. Session 2 included four presentations on digital systems for diagnostics of historic constructions. Session 3 had four presentations on machine learning and pattern recognition approaches for cultural heritage understanding. Session 4 included three presentations on the digital transition in architectural engineering. The Organizing Committee authorized the online presentation of three papers via Microsoft Teams due to last-minute difficulties faced by the authors in reaching the conference location.
In total, VIPERC 2024 successfully hosted 40 participants.
•
We extend our heartfelt gratitude to all the members of the Scientific Committee, reviewers, authors, and participants of VIPERC 2024 for their invaluable support and attendance. A special thank you goes to the IEEE CASE 2024 Organizing Committee, represented by General Chair Prof. Mariagrazia Dotoli, for their warm hospitality and assistance in organizing VIPERC 2024 in Bari, Italy.