Creating 3D virtual sculptures from vision and touch technologies

  • Authors:
  • Antonio Adán Oliver;Vicente Dominguez González;Ricardo Chacón Chacón;Santiago Salamanca Miño;Hector Rodriguez Muñoz

  • Affiliations:
  • Universidad de Castilla La Mancha, Ciudad Real;Universidad de Castilla La Mancha, Ciudad Real;Universidad de Castilla La Mancha, Ciudad Real;Universidad de Extremadura, Ciudad Real;Universidad de Castilla La Mancha, Ciudad Real

  • Venue:
  • Proceedings of the 3rd international conference on Digital Interactive Media in Entertainment and Arts
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

One of the lacks of the current conventional 3D sensor devices is that they are unable to carry out an efficient fusion between overlapped samples and generate complete and realistic reconstruction of an object (Figure 1). The completeness concerns both geometry and colour aspects (G+C). Consequently, obtaining whole (G+C) and precise registration of multiple views of an art piece is an unsolved and important problem which is required in museums and archaeological environments. In this paper we present a computer-human interactive system based on vision and touching technologies that allows us to obtain realistic 3D sculpture models. Following a global-to-local mixture strategy, millions of points with their respective colours, taken by a laser-scanner sensor, are integrated into a unique 3D model. After that, this model can be refined by a sculptor using a force-feedback device in which he can change geometrical, texture and colour properties. The final result is a high-quality restored virtual 3D model which can be exposed in museums or demanded by art researchers. An extensive experimentation is being carrying out on millennium sculptures and pieces of the Spanish National Museum of Roman Art yielding excellent results.