Comic Shading

    The idea behind this project was to model the shading characteristics of cartoon and comic book characters.  

Traditional shading models such as Phong and Gouraud interpolation give a smooth gradation between light and dark

 areas, a technique rarely seen in hand drawn animation.  To give a more comic feel to rendered objects, a discretized 

shading model was developed, based in part on Gooch et al [1].   This shading model performs the standard 

dot product calculations between averaged vertex normals and the light vector, but instead of using this scalar to 

smoothly interpolate the face color between light and dark, the average of the vertex normals is used as a look up into a 

shade table.   This shade table contains a set of bounds, such as: ( Color1, Dot = (0,.2), Color2, Dot = (.2,.8), (Color3, 

Dot = (.8,1) ) .   The face color is then set to the appropriate color.  This method can be though of as simply a 

discretized flat shading model.  The second method utilized to obtain a cartoon shading effect was the use of silhouette

and highlight lines in the image.  The implementation of this was fairly straightforward.  First, all polygon edges for

which one connected face was front-facing and one back-facing was drawn in black to silhouette the object.  Secondly,

all edges for which the normal of the two connected faces was greater than some threshold were also drawn in black. 

The idea here was to find all sharp angles, or creases, in the image to give a visual outline as to the shape of the object.

The following picture demonstrate this technique :

Finally, a brief movie will demonstrate the cartoon shader in animation.