Using 3D Scanning to Analyze a Proposal for the Attribution of a Bronze Horse to Leonardo da Vinci

Abstract

In this paper we present how technologies developed in the context of 3D graphics (3D scanning and image-to- 3D-model mapping) have been used to evaluate in a metrically objective way the plausibility of an attribution hypothesis of a Renaissance artifact. The artifact considered is a small bronze horse (Archeological Museum, Florence, Italy), which was discovered to be very similar to a silverpoint drawing by Leonardo (Leonardo drawing #358, Windsor Royal Library, UK), thus originating an attribution hypothesis. A highly accurate digital replica of this small bronze statue was reconstructed by means of triangulation-based 3D scanning technology. Using image-to-3D-model mapping techniques the Leonardo s drawing depicting a very similar horse has been aligned to the digital 3D model, reconstructing in a virtual manner the ideal vantage points which could have been used to sketch the drawings of the horse (in the hypothesis that Leonardo had drawn it by directly looking at the bronze). The same approach has been also tried with other Leonardo s drawings depicting a similar subject, but none of those was sufficiently compatible in shape with the bronze horse to allow convergence of the image-to-3D mapping process. The approach proposed allows us to give some visually objective evidences about the shape similarity issue, which was the origin of this attribution attempt. At the same time, a purely technical evaluation does not close the attribution issue. Further archival research and expertise will be needed to solve and assess the issue of this disputed hypothesis.