News Archives

Importing Realism into Computer Graphics

August 10, 2005

  • Date: Wednesday, August 10, 2005 
  • Time: 3:00 p.m. 
  • Place: FEC 141

Dr. Marcus Magnor
Max Planck Institute Saarbruecken, German

Expectations on computer graphics rendering quality are rising continuously: whether in flight simulators, surgerical planning systems, or computer games, ever more realistic rendering results are to be achieved at real-time frame rates. And in fact, from the computational as well as algorithmic side, visual realism is within reach of modern PC graphics boards.

With rapidly advancing graphics hardware capabilities, the modeling process is becoming the limiting factor towards realistic rendering. Higher visual realism can be attained only by having available more detailed and accurate scene descriptions. So far, however, modeling 3D geometry and object texture, surface reflectance characteristics and scene illumination, character animation and emotion is a labor-intensive, tedious process. The cost of increasing model accuracy using conventional approaches today threatens to stall further progress in realistic rendering applications.

In my talk, I present an alternative modeling approach. I describe how real-world scenes and events may be acquired from the “real thing”. Given a handful of synchronized video recordings, complex, time-varying scenes and natural phenomena can be modeled from reality to be incorporated into time-critical 3D graphics applications. Photo-realistic rendering quality and truly authentic animations are the result. Besides offering a solution for realistic rendering applications in computer graphics, research into video-based modeling and rendering algorithms also leads to new tools for video editing and may even make possible novel forms of visual media.