Extraction of Three-dimensional Information from Images -- Application to Computer Graphics

This dissertation focuses on the creation of the information used in Computer Graphics. We especially concentrate on the data needed to render images. We do not ask the user to work directly on the object modeling but we rely on her to provide one or several images of the targeted object. These pictures are then automatically analyzed to extract the sought data. From this approach, we expect data more faithful to the original object and a shorter creation time for the user.

Our work is centered on three case studies that have useful applications. We first recover the surface of a matte object from a short image sequence whose viewpoint is moving. We then show how the appearance of a human face is retrieved from a single image and how the extracted data are used to render the original face under a new lighting environment. We end with a technique to capture the hair geometry using multiple images from a fixed viewpoint and a moving light.

We introduce several theoretical and technical contributions that enhance both precision and robustness of the capture. Results are provided to illustrate these improvements.

@PhdThesis\{Par04,
  author       = "Paris, Sylvain",
  title        = "Extraction of Three-dimensional Information from Images -- Application to Computer Graphics",
  school       = "Universit\'e Joseph Fourier",
  month        = "October",
  year         = "2004",
  url          = "http://artis.inrialpes.fr/Publications/2004/Par04"
}