Our approach allows material properties found in 2D art work to be retargeted to 3D geometric models. This work incorporates established principles from the art, computer vision, and rendering communities into a framework for non-photorealistic rendering. Our goal is to leverage art work to obtain effects which may be difficult or impossible to produce with current shading methods. We provide an interface which can be used to tailor the artistic shading model to the aesthetic preferences of a user. Our method allows the user to interactively explore novel viewpoints of 3D models, capturing the light and material properties encoded by artists in 2D imagery.
First, we must appologize to all of these individuals for forgetting to include
this acknowledgements section in the paper (it never got uncommented).
Thanks to Michael Stark, Peter Shirley, Louise Bell, Grue, Richard
Coffey, Susan Ashurst, Matt Kaplan, Jim de St. Germain, and the
members of the University of Utah Computer Graphics groups for help in
the initial stages of the paper. Thanks also to Michael Cohen, James
Mahoney, and Wolfgang Heidrich. Thanks to Viewpoint Data Labs for the
girl model and to University of Utah Alpha\_1 group for the crank
shaft model. Thanks to Mark Harden (www.artchive.com) for the scanned
images of Michelangelo's ``Creation'' and for Cezanne's still
life. This work was supported in part by DARPA (F33615-96-C-5621) and
the NSF Science and Technology Center for Computer Graphics and
Scientific Visualization (ASC-89-20219). All opinions, findings,
conclusions or recommendations expressed in this document are those of
the author and do not necessarily reflect the views of the sponsoring
agencies.