Capturing archaeological reality
3D digitisation of archaeological objects as part of the CARe project
Archaeological objects are three-dimensional artefacts. Nevertheless, they are usually represented in databases, publications and exhibitions by photographs taken from a few fixed angles. For many finds, this is a suitable and comparatively time-saving form of documentation. Three-dimensional recording offers added value particularly where the shape, significance, accessibility or intended use of an object justify it.
The Capturing Archaeological Reality (CARe) project combined the photographic documentation of larger archaeological collections with a focus on the 3D digitisation of selected objects. As part of the second call for proposals under the 'Kulturerbe digital' (eng: digital cultural heritage) funding programme, the State Collections of Lower Austria documented a total of 3,137 objects from the collection areas of Roman Archaeology, Prehistory and Historical Archaeology within a period of one and a half years. Of these, more than 800 selected objects were digitised as high-resolution, photorealistic 3D models and were made available online.
The deliberately ambitious scope of the project was not intended solely to produce a large number of models. The aim was to test, coordinate and standardise a high-quality workflow on a large scale under real-world conditions. At the same time, the technical, methodological and personnel requirements were put in place to enable the three-dimensional capture of suitable objects beyond the duration of the project.
A large number, but no random selection
Despite the ambitious project objective, objects were not selected at random simply to reach the target number as quickly as possible. The focus was on finds for which a specific need or intensive subsequent use of the date generated was anticipated.
Objects-specific reasons, such as a complex shape, a distinctive surface, a mechanism that is difficult to understand, or fragility, could justify three-dimensional digitisation. Equally, historical and scientific significance, specific questions arising from research and restoration, and planned use in exhibitions, educational projects or digital outreach were taken into account. Objects that were difficult to access and frequently requested were also considered.
Together, these criteria formed the professional framework for the selection process, though they did not have to apply equally to every object. Both well-known highlights of the collection and finds that had previously received less attention – which only came to the forefront through targeted research in storage areas and collection of finds – were considered.
The decisive factor was therefore not merely technical feasibility, but the demonstrable benefit of a model for documentation, research, conservation or outreach. These criteria will continue to form the basis for the selection of suitable objects in the future.
From spatial recordings to a convincing model
Objects can now be captured in three dimensions using relatively accessible and cost-effective methods, such as smartphone apps or handheld scanners. However, CARe deliberately chose not to use the quickest method for producing simple models. The aim was to achieve a reproduction that was as true to life as possible, in which not only the spatial form but also colours, surfaces, traces of workmanship, decorations and damage remain recognisable.
The work is by no means complete once the image sequences have been captured and an initial model has been calculated automatically. It is particularly evident from some 3D models published on the internet that capture and automatic calculation alone do not guarantee a convincing result. The subsequent post-processing is at least as important.
This involves correcting geometric errors, such as closing gaps, reconstructing originally open cavities and removing digital noise or seemingly free-floating particles. Differences in colour and lighting on the surface must also be balanced out.
At the same time, the calculation process initially generates large volumes of data. These must be optimised so that the models can be displayed smoothly in a web browser or on media stations without losing their detailed and high-quality overall appearance. It is only through this technical and creative refinement that the automatically generated initial model is transformed into a visually convincing and reliably usable digital replica.
In order to meet these quality standards for more than 800 objects within the project timeframe, recurring work steps were standardised and, in some cases, automated. Surveying, immediate quality control, calculation, post-processing, scientific documentation and publication were closely integrated. This made it possible to reduce unnecessary manual repetition without comprising on quality.
The survey procedure and the combination of focus stacking and photogrammetry used in the project have already been reported on in more detail in the activity report 'Im Bestand 2025' (eng. 'In the 2025 inventory') by the State Collections of Lower Austria.
When the function only becomes apparent through movement
A bangle from the Prehistory and Historical Archaeology collection, which was digitised as part of the project, illustrates the additional insights a 3D model can provide. It features an unusual opening and closing mechanism. From individual photographs alone, it is difficult to understand how the piece of jewellery was opened, put on and closed again.
The 3D model not only allows the object to be viewed from different angles, but also forms the basis for an animation of its movements. The digital replica thus not only documents the object's appearance, but also makes its original function and handling comprehensible.
The published models can be freely rotated, zoomed in on and viewed from angles that would not be accessible in a display case or in a traditional photograph. In exhibitions, for example, they can be used via media stations or QR codes to reveal the reverse sides, undersides and small details.
Furthermore, the models can serve as a basis for 3D prints or digital reconstructions. Reproductions could, for instance, temporarily fill a gap in an exhibition on the event of loans or restoration work. In the case of finds that have survived in fragments, it is possible to illustrate their likely original appearance, provided that the preserved components and hypothetical additions remain clearly distinguishable from one another.
An infrastructure for additional collections
The State Collections of Lower Austria hold around six million objects. Only A fraction of these require three-dimensional documentation. However, the methods developed as part of the project will make it possible in future to record further suitable finds on an as-needed basis and to a high visual standard.
For the implementation and methodological validation, the project collaborated with the Centre for Museum Collection Sciences at the University for Continuing Education Krems. As part of this collaboration, a specialist in 3D digitisation was employed to carry out the technical capture and post-processing of the models and to train staff at the State Collections of Lower Austria in the methods used whilst the work was in progress.
The more than 800 models produced are therefore not merely the outcome of a project. They also represent a tried-and-tested workflow, permanently available specialist knowledge and a technical infrastructure that can be used to document further objects in three dimensions according to scientifically sound criteria.
3D digitisation replaces neither the original nor traditional object photography nor scientific description. When used correctly, it complements these forms of documentation by adding the spatial quality of an object. Its particular value is evident where it makes form, surface, function and interrelationships easier to understand and opens up new possibilities for research and outreach.