I am Titien, a French archaeologist specialized in Roman architecture and decoration, and involved in Digital Humanities matters and technologies, more particularly in digitization, 3D modeling, and numerical simulation. I perform these activities and conduct research within Artifact Imaging, a brand-new company that produces high quality scans, 2D and 3D scientific graphic documentation for analysis, mediation, enhancement, and preservation of heritage, aiming to provide innovative digital solutions for players in cultural heritage, research, and the art market.
I have been involved in the field of digital development and the use of new technologies in archaeology since 2008, notably through my participation in the ANR program OrAG (Architectural Ornamentation of the Gauls). OrAG aimed to develop a participatory and interactive database to highlight the lapidary collections kept in museums and considered the contribution of digital technology to the survey and to the representation of the pieces. It is within this framework that I began to focus on the digitization of cultural heritage, on the techniques and their benefits but also on the considerable contribution that digital methods represent for collections that are sometimes unrecognized, poorly valued, or in danger. I then participated in various working groups, roundtables, and conferences.
From 2014 to 2018, I coordinated the team of the Institute for Computing and Data Science of Sorbonne University responsible for digitizing and enhancing archaeological collections. Participating actively in the development of Sorbonne University’s digital archaeology team, I implemented the 3D acquisition and processing protocols that we used and designed digital rendition tools and analysis solutions (interactivity and immersive nature of models, conception of web applications and digital rendering software, non-invasive digital restoration, etc.). In this context, I daily juggled between traditional archaeological activity and the use of 3D as an essential component of my research.
The project of virtualization of the archaeological collection of Orange is a perfect example of the contribution of the digital to a complex collection whose archaeological and architectural reconstructions are difficult. This collection is characterized by incomplete decorations on scattered pieces conserved in different places. The museum’s proposal to put digitized collections online is a natural follow-up to several research projects carried out by the Sorbonne University digital archaeology team.
Mainly, I use 3D for heritage preservation, analysis, and valorization. In terms of analysis, I undertake, for example, volumetric comparisons and I detect and interpret evanescent technical traces to clarify, confirm or invalidate a traditional analysis, to identify the manufacturing tools and reproduce a process of construction or destruction. Then, I can make virtual reconstructions or non-invasive restoration of the missing parts or the original colors. We are already in the field of digital valorization, which also includes virtual visits, temporary and thematic exhibitions, or reconstruction of the archaeological context, among other things. Finally, 3D is a catalyst in the production of traditional graphic documentation (cut and section, plan and elevation, unrolled decorations of convex support, ortho-images and ortho-photographs, etc.). Going through digitization ensures an absolute accuracy to these graphic documents, which aids in the analysis and the interpretation of a subject and in its appreciation.
Some examples below: a plan of the excavations of the sanctuary of Gabii, Italy; a traditional orthoview of an architectural piece from the decoration of the theatre of Orange; a restoration of the gold leaf covering and the missing parts of a Canaan goddess figurine.
Concerning the archaeological collections and the cultural heritage of the city of Orange, providing digitized objects on the museum’s Sketchfab account is only a starting point. We consider a whole digital strategy—in connection with the project to create a new museum—that will include the digitization of built heritage (among others the ancient theater and triumphal arch, both UNESCO World Heritage sites), and the virtual restitution of the mausoleums of Fourches Vieilles. These 3D models will be enriched with the existing graphic and scientific documentation to give them an educational dimension justifying the proposals, which will be done in consultation with the researchers, archaeologists, and heritage architects working on these issues. The multi-scale dimensions will be illustrated by 3D digital models that place these monuments within the known ancient city. And, of course, each of these productions will be made available on the Sketchfab account of the museum. Finally, beyond the webcast, innovative solutions will be proposed to offer visitors a resolutely modern experience and new mediation tools.
Tools and Workflow
I do photogrammetry exclusively for digitization, which remains the most flexible technique and is able to adapt to any type of subject and any type of environment. The acquisition process obviously depends on the type of object, its dimensions and physical constraints. It can be done with a traveling studio or a lightbox and is preceded by a protocol set up specifically for the determination of lighting and the number of shots needed. But photogrammetry can also be done in a more manual way when faced with emergencies in the field, where work is often subject to tight timing or difficult-to-access architecture, terrain, or objects. Such is the case for the example below, the statue of Augustus placed on the wall of the theatre of Orange. I took advantage of the scaffolding installed for its restoration but I had to deal with the lack of accessibility and short distance to the sculpture.
In terms of equipment, I use an APS-C camera, which is the best compromise to avoid heavy photos and long calculation times. Obviously a tripod and a remote control are used as far as possible and the lens depends on the size of the object, but I like to stay between 35mm and 50mm. The eventual artificial lighting depends on the environment, on the method and on the reflectivity of the object.
Photogrammetric processing is then performed in Agisoft Metashape and scaling is usually done by using targets whose distance measurements are listed (stuck on the turntable when I use the lightbox or on rulers prepared and arranged around the object).
When the objects being photographed can be easily handled, I am able to create complete models of them. I create two photo sets, turning the object halfway through, and gluing the 2 chunks together during the calculation. This results in a harmonious sphere of cameras distributed around the model, ensuring the complete scan.
Once processed, I import the models into Blender to retopologize, optimize, and improve textures. I have become used to always building my models as true and close to the original as possible to preserve a maximum amount of archaeological data, then to extract lighter doubles according to their intended use—web, visualization, or analysis.
It is also in Blender that I work on restitution/reconstitution, scenography, and rendering calculations. The last step of my work is to meet specific expectations depending on the project: production of orthonormal images and plans; detecting, interpreting and highlighting particular traces, such as the interpretation of tools used, traces of antique restorations and repairs, or traces of persistent polychromy; building digital anastylosis; creating digital collections, virtual exhibitions or videos; and many other things….
The main obstacles that have to be faced in digitization work are all related to the nature of the collections, the associated research issues, and the purpose of the digitization. That is to say, to process important batches in record time, to preserve archaeological data essential to the analysis. Therefore, it is critical to preserve optimal fidelity in terms of volumetry, details, and dimensions, and to enable handling multiple heavy 3D objects in the same environment. Hence the need to set up reliable acquisition and data management protocols upstream, to carry out a very high definition scan and to systematically process scanned objects to optimize them according to the use and the end user. These steps are systematically dependent on the collaborators or partners and their interest, to meet their desires in terms of analysis, valorization and publication, multi-scale visualization, etc.
Getting Started in 3D
My advice would be to take the time to do it right. When it comes to photogrammetry, it is all about good pictures, a good mesh, beautiful maps and textures. In the section above, I raised the point about identifying the purpose of the initiative, because we can face many obstacles and errors that can be avoided if they are anticipated. When it comes to cultural heritage, the main objective is twofold and arises directly from archaeological and patrimonial concerns: to record a datum that will be saved and can be studied; to take advantage of these data to enhance a science, a collection, and a cultural heritage by making it accessible to as many people as possible.
The Future of 3D in Heritage
Researchers and fieldworkers in archaeology have already incorporated the advantages of digitization—recording, support for traditional documentation, virtual reconstructions, etc.—and now the growth of digitization techniques is only the logical continuation of pre-existing concerns, illustrated by digital terrain models, GIS, topography, and geo-magnetic surveys, for example. It is obvious that these new tools will become systematic on archaeological excavations, for a science which, by nature, remains most often invasive and destructive. The next great advance will be managing, exploiting, and valorizing all this data by making it accessible and shared, a task which also responds to the desire of the public for access to heritage. In this sense, museums, foundations, and institutions should all consider setting up a digitization strategy for their collections. Sketchfab is doing a great job in providing heritage stakeholders and archaeology researchers with a platform to share their work, and customized solutions that already meet these expectations of data sharing and accessibility.
Favorite Projects on Sketchfab
A lot of models are great and it is not easy to choose one from all, but I really enjoyed the project led by Virtual World Heritage Lab at Indiana University in partnership with the Uffizi for the completeness they have shown on a remarkable collection. I am also very interested in the Monte Albán Digital Archaeology Project, which highlights the notion of multi-scale, ranging from the main plaza in its entirety, to the object itself, through to the architectural structures, all of which I like.
Finally, I was impressed by the quality of the models produced by the Museo Arqueológico Nacional. A Greek pottery collection, black and red figure, with great rendering and meshes (anyone working on the digitization of cultural heritage should take a look at the bump, roughness, AO and cavity mapping on their models). If I had to choose only one to illustrate what is a good job of digitization, valorization, and dissemination it would be one of the three groups mentioned here.