Visual encoding of a 3D virtual reconstruction's scientific justification: feedback from a proof-of-concept research
3D virtual reconstructions have become over the last decades a classical mean to communicate about analysts’ visions concerning past stages of development of an edifice or a site. However, they still today remain quite often a one-shot output, neither reusable, nor expressing visually the diversity of the evidence behind the reconstruction. A 3D virtual reconstruction - obtained by means of a synchronic or diachronic study - is a hypothesis, but also a technical work, and ultimately confronts researchers with the challenge of finding a compromise between interpretation and assertions, even within the graphical encoding of a 3D model.
This research is based on a methodological proposal: introducing justification matrices that associate “indicators” to the components of a 3D virtual reconstruction, that formalise in a synthetic way an assessment of plausibility. These matrices allow assigning individual 3D objects or groups of objects a quantitative evaluation of their plausibility by crossing four criteria: shape, dimensions, existence, and position.
The experiment led to the creation of a proof-of-concept prototype allowing a user to interact in real time with the graphic appearance of each object, represented in a 3D reconstruction, according to the values of its justification matrix. The approach is applied to four phases of the evolution of Marmoutier abbey's hostelry (corresponding to four synchronous states - 12th, 13th, 15th and 18th centuries), and re-uses 3D virtual reconstructions produced several years ago for communication. The paper positions and discusses the three families of issues the research intersects (knowledge modelling, visual encoding and 3D content reuse), presents the case study, details the services offered by the prototype and assesses lessons learned and limitations. The experiment should be seen as an attempt to reason about heritage data sets by drawing on practices from the InfoVis field and is thought of as a call to design and discuss in the scientific community news ways of visualising architectural data, information, and pieces of knowledge.
- SESAMES – Sensemaking and Spatialisation for multi-granular heritage artefacts : 3D annotations, sonification. and formalisation of inferences ANR-18-CE38-0009
- Agence Nationale de la Recherche