Today, there is a tendency to mix real images with computer-generated images at the point of recording so that producers can use a three-dimensional (3D) result to judge whether a scene will appear in the final production and to guide actors, as well allowing the results of filming to be directly inserted into a traditional computer graphics production chain. To satisfy this demand, a large body of research has focused on multiview reconstruction. This chapter focuses on silhouette-based reconstruction and its improvement. Following a detailed description of the different stages in this method, its implementation in an industrial context is examined. The challenge of temporally structuring multiview reconstructions lies in converting a sequence of meshes into a normalized representation when animating characters. Finally, the chapter presents an overview of multiview reconstruction analysis techniques to extract temporally stable semantic information to facilitate their integration into the computer graphics production chain.