9

1732 Chapter 16: Material Editor, Materials, and Maps
Method group
The Method group lets you cho ose which
coordinate to use on the normals. These controls
are the same as those in the Projection Options
dialog (page 3–165).
•
Tangent—(The default.) Project at a tangent to
the target object’s surface.
This is the metho d to use for objects that both
move and deform, such as animated characters.
•
Local X YZ—Project using the object’s local
coordinates.
This method can be used for stationary or
moving objects, but not for objects tha t deform:
if the object deforms, the projection will appear
incorrectatsomeframes.
•
Screen—Project using screen coordinates; that
is, flat projection in the Z axis. X is horizontal,
increasing in a positive direction to the r ight;
Y is vertical, increasing in a positive direction
upward; and Z is perpendicular to the screen,
increasing in a positiv e direction toward the
viewer.
This metho d is useful mainly for stationar y
objects seen only from a sing le angle; for
example, a statue seen through a window.
•
World—Projec t using world coordinates.
This is useful mainly for objects that don’t move
or deform; otherwise, a moving object with
world-projected normals will appear to “swim”
through the texture.
Came ra Map Per Pi xel Map
Material Editor > Maps rollou t > Click a Map button. >
Material/Map Browser > Camera Map per Pixel
The Camera Map Per Pixel map lets you project
a map from the direction of a particular camera.
It is meant as an aid to 2D matte painting: You
can render a scene, adjust the rendering using an
image-editing application, then use this adjusted
imageasamattethatisprojectedbackontothe
3D geometry.
Tip: Final rendering can be slow. Script-driven
network rendering (page 3–173) can help improve
performance.
Limitati ons
The C amera Map Per Pixel does not handle these
situations:
•Animatedobjects.
The projection does not use UVW mapping.
•Animatedtextures.
• Occlusion based on a Z-depth channel is
handledinalimitedwayonly.
Procedure
To use Camera Map Per Pixel:
1.
Create the 3 D model.
2. Set up a camera.
3. Set up the rendering resolution you want.
To get go od results, the plate should be at least
2K pixels; 3K to 6K, or higher is recommended.
4. Render the scene to an editable image format
such as TIFF (page 3–634).
5. Render the scene again, this time to a format
such as RPF (page 3–631) that has a Z-depth
component. Make sure the Z option is turned
on.
6. Use an image-editing application to make
changesyouwanttotheeditableimage.
7. Apply Camera Map Per Pixel to the diffuse
component (page 2–1498) of the geometry on
which you want the matte to appear . U se these
settings:
• SetCameratothesamecamerayouusedfor
the renderings.