FINAL PROJECT - Now that you have vast experience with Renderman, Renderman shaders, and MTOR, you have one final two part project to complete the semester.

For part 1, are asked to select or create a photograph of a complex real world scene.  You are then asked to model a moderately complex CG element appropriate for your scene.  Then using Renderman, light and render this element.  Then composite this rendered element (or elements) into your real world scene.  The goal is to seamlessly fit your rendered CG element into the scene such that it appears to have been there when the scene was originally photographed.  You need to have a believable object, with correct camera position, lighting, shading, colors, shadows and reflections that fit into the original scene.

For part 2, you are asked to photograph a single relatively simple real world scene from two points of view.  Your scene may be a carefully controlled studio setup or some existing environment.  Your scene should be visually interesting and present some challenges to your rendering skills.  You are to create a CG model of this environment and using Renderman carefully render images of this model to exactly match the original scene photographs.  The goal is to match your rendered scenes to the real world scenes as closely as possible.

You need to turn in for each part:

1) A README file.  Explain how you went about recreating your real world scenes.
2) The source code for your shader(s).
3) The compiled Renderman shader(s).
4) The Maya and/or RIB models (including lighting) for your scenes.
5) Any texture, displacement, ... maps used
6) For part 1 - at least three (3) images (jpeg). The original scene, the rendered CG element(s) only, and finally the CG element(s) composited into the original scene.
7) For part 2 - four (4) images (jpeg).  Two original scene photographs and two CG scenes created using Renderman.  The rendered image pairs should "exactly" match the original scene photographs.

Please use appropriate commenting and intelligent naming of variables so that the shaders can be easily deciphered.