DIGITAL HERITAGE 2015
Special Session Proposal
The role of similarity in the re-unification, re-assembly and re-association of 3D
artefacts
Thematic area: The workshop fits primarily in the Analysis and Interpretation area and also in the Digital
Heritage Projects and Applications.
Proposers:
Michela Spagnuolo is Research Director at CNR-IMATI-GE, where she is leading the activities on Advanced
techniques for the analysis and synthesis of 3D shapes. Her research interests include geometric and
semantic modelling of 3D objects, computational topology for the analysis of shapes, method for the
evaluation of similarity at the structural and semantic level. She authored more than 130 reviewed papers,
is associate editor of international journals in Computer Graphics (currently, The Visual Computer and
Computers&Graphics). She is member of the steering committee of Shape Modeling International, EG
Workshop on 3D Object Retrieval and EG Workshop on Graphics and Cultural Heritage. In 2014, she was
nominated Fellow of the Eurographics Association.
Bianca Falcidieno is a Research Director of the National Research Council (CNR) of Italy. She has been
leading and coordinating research at international level in advanced and interdisciplinary fields (such as
computational mathematics, computer graphics, multidimensional media and knowledge technologies),
strongly interacting with industrial, cultural and social application fields. She coordinated several national
and international projects and was in charge of various international commitments, including editorial tasks
in international journals (currently, Computers&Graphics and Graphical Models), organization of
international conferences and workshops as the Chair or Co-chair and participation in international
conference program committees. Bianca Falcidieno is the author of over 200 scientific refereed papers and
books. She is a EUROGRAPHICS fellow and for the 80th CNR anniversary, she was included in the 12 toplevel female researchers in the history of CNR.
Contact information:
Institute of Applied Mathematics and Information Technologies - National Research Council– Genova, Via
De Marini, 6 - 16149 Genova (Italy)
Ph. +390106475671, Fax: +390106475660 - e-mail: {michela.spagnuolo, bianca.falcidieno}@ge.imati.cnr.it
Workshop Synopsis :
Digital manipulation and analysis of tangible cultural objects has the potential to bring about a revolution in
the way classification, stylistic analysis, or refitting of fragments is handled in the cultural heritage area: 3D
modelling, processing and analysis are now mature enough to allow handling 3D digitized objects as if they
were physical, and semantic models allow for a rich documentation of many different aspects of artefacts or
assets of any complexity, as well as of contextual information about them.
In this context, the workshop is focusing methods for analysing, presenting and documenting digital
cultural assets, with focus on research challenges related to the re-unification, re-assembly and reassociation of artefacts. Re-unification is the process of discovering parts of the same object held in
different collections and evaluate if and how they could fit together. Re-assembly consists in digitally
recreating an historical artefact by the set of its fragments. Re-association of objects allows researchers to
look for new understanding and insights into the movement and links between different communities on
the basis of similar artefacts found in different locations. Similarity evaluation is underlying most of these
challenges, as the ability to reason on the several and diverse artifact properties, which may relate to
geometric attributes (e.g., spatial extent, aspect), to colorimetric properties (e.g., colour, texture), to
specific traits that fragments exhibit (e.g., decorations), or to metadata documenting the artefacts.
Emphasis will be given to the role of shape analysis for fragment analysis, which is a specific area where the
digital setting may support a safer and enhanced manipulation of fragile and precious artefacts. The vast
majority of archaeological objects are discovered indeed in a fragmentary state; then, the extent of
archaeological research possible is hampered by the poor state of preservation of these objects, and much
effort and time is expended in trying to cluster and re-compose these fragments as accurately and
completely as possible to reconstruct the original object representations. This process is further
complicated when the different parts of these artefacts have become dispersed amongst a number of
different collections, whether public or private. In some cases, the owners of these collections prove very
reluctant to give up physical artefacts to others, but usually have no objections in releasing 3D surrogates of
these items. Here, the need of analysing incomplete and degraded objects will be matter of discussion and
state-of-the-art contributions will be presented. Complementarily, re-association instead opens up new
perspectives on how the analysis of geographically dispersed cultural objects might be analysed, thanks to
novel means of processing, comparing and matching shape, stylistic traits and any other semantics-driven
information that could be integrated in a comprehensive and multi-modal documentation of the digital
assets.
Programme of the workshop (ordering of the talks subject to changes)
The workshop programme brings together researchers and professionals in the Cultural Heritage area and
museum curators, all contributing to the workshop with different perspectives on the various facets of
similarity assessment for the re-unification, re-assembly and re-association in cultural heritage.
Introduction to the workshop: The role of similarity in the re-unification, re-assembly and reassociation of 3D artefacts, Bianca Falcidieno, Michela Spagnuolo, CNR-IMATI
1
The application of semantic techniques to CH artifact description, contextualisation and
discovery of associative relationships between artifacts. Paul W Walland
The vast majority of archaeological objects are discovered in a fragmentary state, and the poor state of
preservation of these pieces further hampers the extent of archaeological research it is possible to do on
them. In today’s digital era it is becoming possible to render such fragments as virtual 3D objects, to
manipulate them, match them and 3D print them as an alternative to handling or transporting the physical
objects. Indeed, the use of virtual representations enables workers and researchers to work together from
different geographical locations on artifacts from a number of different collections simultaneously without
recourse to physical transportation of the objects themselves. In this paper we will present a new approach
to managing digital representations of artifacts and artifact fragments in which semantic annotations are
used to capture their physical and abstract (non-physical) characteristics, and semantic reasoning is used to
induce relationships between them. By this means we aim to show that related cultural heritage artifacts
can be identified and matched more efficiently than at present, and by extension of this approach show that
more abstract relationships between objects, such as provenance, historical association, geographical origin
or style can also be discovered, thus adding to historical knowledge such as migratory or trading patterns in
different periods of history.
Paul Walland is Innovation Director at the IT Innovation Centre, part of the University of Southampton’s
School of Electronics and Computer Science in the UK. He is responsible for a portfolio of projects in the
areas of media, social media and knowledge management, with additional interest and responsibilities in
the generation of innovative applications based on research outputs. Prior to joining IT Innovation Paul
worked for more than 25 years leading industrial research groups in topic areas ranging from optical
systems to media and broadcast technology. He has chaired workshops, conferences and technical working
groups and has published widely at international conferences and in technical journals.
DIGITAL HERITAGE 2015
Special Session Proposal
2
The Salamis Terracotta Statues – a multidisciplinary research approach. Sorin Hermon, Vera
Moitinho de Almeida - STARC, The Cyprus Institute
In this presentation we will introduce a multidisciplinary research approach for a comprehensive
investigation of a large corpus of terracotta fragments (second half of 7th century B.C.) discovered at the
archaeological site of Salamis (Cyprus). Fragments of these statues are presently exhibited or stored in
various public and private museums and collections scattered around the world. The goals of our research
include: (1) archaeological description, analysis, classification and documentation of each fragment; (2)
integration of digital techniques with analytical chemical analysis; (3) virtual restoration of these statues as
much as possible. This is to say, the re-assembly, re-association and re-unification of the Salamis terracotta
statues corpus. The long-term goal of such research aims at gaining new insights into the related chaîne
opératoire and social recontextualization, for a better understanding of past societies. To this end, we
propose an approach based on the visual/non-visual, qualitative/quantitative, geometric, colour and
material (pigments and clay) characterization and analysis of every fragment. Such an approach is based on
the use of portable non-destructive methods and techniques: from 3D scanning, to digital microscopy, x-ray
fluorescence (XRF), spectrocolorimetry and multispectral imaging. A preliminary and direct outcome
consists in a 3D digital corpus of these fragments, which will be archived into a digital repository at the
Cyprus Institute and to be linked with The British Museum online database.
Vera Moitinho de Almeida is a post-doctoral fellow at STARC, The Cyprus Institute. She obtained her Ph.D.
from the Universitat Autònoma de Barcelona (UAB), focusing on technological and functional analysis of
archaeological objects, using 3D models and Reverse Engineering processes. She has a M.Sc. in Prehistoric
Archaeology (UAB), an interdisciplinary M.Sc. in Multimedia Technologies (University of Oporto), and a BA in
Fine Arts. She is the Cyprus representative in the COST action “Colour and Space in Cultural Heritage”
(TD1201). She is co-author of more than thirty scientific publications in the field of 3D applications to
research on tangible cultural heritage assets.
______________________________________________________________________________________
3 Addressing Similarities for Predictive digitisation, re-assembly and completion of 3D Cultural
Heritage Objects in the PRESIOUS project. T. Theoharis, I. Pratikakis, G. Papaioannou, T. Schreck, D.
Rieke-Zapp, A. Andreadis, F. Arnaoutoglou, R. Gregor, P. Mavridis, M. Savelonas, k. Sfikas, I. Sipiran, K. Vardis
The talk will address two key objectives of the PRESIOUS project that invoke similarities. The first objective
concerns how similar reference models are exploited to allow gradual shape prediction from partially
digitized objects, resulting in reduced 3D scanning times for specific settings. The second objective regards
automated, predictive reconstruction from fragmented CH objects that goes beyond reassembly by
proposing the synthesis of missing parts using geometry auto-completion, thus aiding the physical and
virtual restoration process.
Ioannis Pratikakis is Assistant Professor at the Department of Electrical and Computer Engineering of
Democritus University of Thrace in Xanthi, Greece. He received the Ph.D. degree in 3D Image analysis from
the Electronics engineering and Informatics department at Vrije Universiteit Brussel, Belgium, in January
1999. In March 1999 he joined IRISA/ViSTA group, Rennes, France as an INRIA postdoctoral fellow. From
January 2003 to June 2010, he was working as Adjunct Researcher at the Institute of Informatics and
Telecommunications in the National Centre for Scientific Research “Demokritos”, Athens, Greece. His
research interests lie in image processing, pattern recognition, vision and graphics, and more specifically, in
document image analysis and recognition, medical image analysis as well as multimedia content analysis,
search and retrieval with a particular focus on visual content. He has published more than 150 papers in
journals, book chapters and conference proceedings in the above areas. He has participated in more than 20
national and international R&D projects. He has served as member of the Board of the Hellenic Artificial
Intelligence Society for the period 2010-2014, he is Senior Member of the IEEE and member of the
European Association for Computer Graphics (Eurographics). In 2015, he co-organised the 8th Eurographics
workshop on 3D Object Retrieval (3DOR 2015).
4 What do thirty-one columns say about a thirty-second? Livio De Luca
With a strong technological presence, the study of the built heritage is facing a problem of “information
overload.” Indeed, this strong technological presence fails to strengthen representation in its role as a
vehicle of knowledge. Confronted with the intelligibility deficit, this presentation focuses on a way for
reading morphological features of an artifact by using a bottom-up approach: the meaning of elements (i.e.,
their semantic layouts) come from a statistical analysis of some main morphological features of a collection
of shapes. The idea is to rely on data accumulation to render apparent high-level semantic structures from
the comparative analysis of common low-level geometric features. The introduced principles will be
illustrated by the study of thirty-one romanesque columns of the cloister of the abbey of Saint-Michel-deCuxa.
Livio De Luca is research director at CNRS and director of the MAP (Models and simulations for architecture
and cultural heritage) laboratory. He received his PhD in engineering in 2006 at the “Arts et Métiers Paris
Tech” and his HDR (habilitation) in Computer Science in 2012 at the Aix-Marseille University. Member of
several international scientific committees for the digital documentation of cultural heritage, his research
activity focuses on digital surveying, modeling and representation of architectural heritage as well as on the
development of semantic-based systems for describing, analyzing, documenting and sharing digital
representations of heritage buildings.
5
Scan4Reco: characterization of artworks materiala through photometric and microscopic
analysis, Anastasios Drosou, Dimitrios Tzovaras, Claudia Daffara, Andrea Giachetti, Carlo Reghelin, Enrico
Gobbetti, Ruggero Pintus
The Scan4Reco project aims at delivering a cost-efficient, portable, integrated system, based on multi-modal
and multi-discipline, modular, scalable and open-architecture extendable platform that will be able to
provide multispectral scanning of a great variety of cultural asset (e.g. wall-paintings, painting, metallic
objects of various sized, carved marble, statues, etc.) non-destructively. This system will be not only used for
material identification, stratigraphy revealing and automatic, accurate digital 3D representation and
reconstruction of the object in its original state, but also for the automatic inference of both previous states
(i.e. restoration) and forthcoming state/shape. Specific tasks of the project will aim at characterizing
material surfaces at a microscopic level using reflectance analysis and microprofilometry. This material
characterization can be clearly exploited for the re-unification, re-assembly and re-association of 3D
artefact.
Andrea Giachetti is Associate Professor at the Dept. of Computer Science of the University of Verona, where
he focuses on image, video and sound processing. He received the Ph.D. degree in physics from the
University of Genova, Italy, in 1993. Since 1997 he has worked with CRS4 (Center for Advanced Studies,
Research and Development, Sardinia) and since 2006 he has been Associate Professor at the University of
Verona, Italy. His research activity is mainly focused on Image Processing and Computer Vision. He is author
of more than 100 publications in international journals and conference proceedings.
6 3D artifacts similarity based on the concurrent evaluation of heterogeneous properties. Silvia
Biasotti, Andrea Cerri, Bianca Falcidieno, Michela Spagnuolo
The analysis and comparison of shapes in the Cultural Heritage domain requires the evaluation of similarity
among real-world 3D artefacts, which may differ greatly in terms of shape, style, color, material and other
DIGITAL HERITAGE 2015
Special Session Proposal
attributes. It is therefore necessary to develop computational methods which are flexible enough to cope
with the variety of similarities that professionals in the Cultural Heritage area could require for their
analysis. In this context, we will discuss a methodology that is able to concurrently evaluate heterogeneous
properties, such as geometric and photometric aspects. The geometric description is based on a statistical
technique to select properties that are mutually independent; the photometric information is handled
according to a topological perspective, and complemented by the analysis of colour distribution. The
outcome is a mixed description of each 3D artifact, which is used to derive a similarity measure between
objects.
Andrea Cerri is currently reseacher at the CNR-IMATI. He received the Ph.D. degree in Mathematics from
the University of Bologna in 2007. From then on, he has been a research fellow at ARCES Centre of
Excellence (University of Bologna) and a project assistant at the Pattern Recognition and Image Processing
group (Vienna University of Technology), until joining CNR-IMATI. His research interests include geometrical
topological methods for shape analysis and comparison, and related applications in computer vision,
computer graphics and pattern recognition.
7 Title [Barry Norton - British Museum - title and authors] TO BE CONFIRMED
Time frame for workshop:
Half-day, approximately 3 hours, plus coffee break.
1 introductory talk, 7 confirmed talks, one pending. The time slot planned for each full presentation is
25min, questions included.
Target audience:
The themes of the workshop covers some crucial aspects of Digital Heritage and are potentially interesting
for all the audience attending the conference. In particular, CH curators and professionals could learn about
the most recent developments in the field of shape analysis and semantic description of 3D digital artefacts
and contribute to the discussion with their own experience. Also researchers in computer science could
take part to the workshop and share their interests and approaches to the geometric and semantic
processing of digital shapes. The organizing team have a long-standing expertise in the organization of
workshop and conferences. Of particular relevance with respect to the proposal topics are:
 3DOR: since 2008, Michela Spagnuolo is member of the steering committee of the EG Workshops on 3D
Object Retrieval, whose topics heavily related to the workshop themes. She has been acting as
workshop chair and programme committee chairs several times within this workshop series.
 Learning Cultural Heritage by Serious Games, workshop at Digital Heritage 2013: half-day workshop in
the past edition on DH supported by the EU NoE GALA. We portrayed the recent proposition of serious
games in the cultural sector from a multi-disciplinary perspective: in fact, the diversity and the
complexity of the field have to be matched with the new technologies and the pedagogical aspects.
 SGP 2013: the Eurographics Symposium on Geometry Processing 2013, a premier venue for research in
the manipulation of geometrical information. Bianca Falcidieno and Michela Spagnuolo have been
serving as conference organizers in 2013.
 SMI: Bianca Falcidieno and Michela Spagnuolo are in the steering committee of the international
conference on Shape Modelling and Applications - SMI, since its injection. They have been serving as
conference chairs or programme chairs several times during the years.
Expected number of participants: Based on past experiences, approx. 60-70.
Equipment needed: PC, LCD projector.
Budget:The workshop does not require any budget. The organisers and invited speakers are self-financing
and no additional fee is required to the participants.