Le patrimoine in silico. Exemple de la cathédrale d’Amiens
Résumé
Several digital heritage projects for the creation of 3D models of historic buildings have emerged during recent years. Such projects have been stimulated by some significant advances in digital surveying technologies and in data processing. The wider diffusion of these techniques and several other methodological advances, notably in computer vision and photogrammetry, have also contributed to this phenomenon. The goal of the E-Cathedr@le programme, developed by our research group and centred on Amiens cathedral in France, is to keep a record of a historic monument which is as accurate and reliable as possible, and to enrich this record with textual information providing the basis for multidisciplinary research. While the objective is to cover the totality of the monument, both inside and out, and to take high-resolution and photometrically-realistic measurements, each of these criteria has its own scientific pertinence and an important impact on the modus operandi and on the quantities of data to be recorded and managed. These challenges forced us to design new adapted modalities of data capture. The programme was launched in 2010 and is expected to last 15 years. It has already allowed for the digitization of several other monuments in France and in Morocco, thanks to a bilateral research project. These digital models have been shared and widely used for multidisciplinary research in computer science, robotics, art history and archaeology. In this paper, we describe the E-Cathedr@le programme and we present its main results. In particular, we show how our 3D model of the monument allows for the easy creation of accurate plans and cross sections of the architecture. This in turn allows for the body of the building to be visualised in an innovative way. We conclude the paper with a discussion of the lessons learned during the first eight years of the programme, and of the main challenges and opportunities involved in digital heritage projects.