Taken as a whole, both the marschner model and the kajiya kay model are both robust methods for rendering hair and fur. The simplicity of the kajiya kay might be preferred for when performance is necessary such as in game. By proposing hair as a collection of cylinders, we can use an anisotropic specular reflection.
On the other hand, Marschner Model is a slightly more involved method. It is based on how hair naturally scatters as may have multiple colors and textures.
muuncakez
Is there any difference when rendering hair that has been styled quite differently? By quite differently I mean braids vs straight hair. Or curly vs straight. And then curly in a bun vs straight in a bun? My first instinct is to think yes, especially since curly hair interacts with its environment differently to straight hair. And then how does one apply this to hair that is up in a bun? Are bun styles just a sphere with hair texture and BRDF? Or a braid is just a manipulated polygon with various volumes to represent each "chunk" of the braid? (referring to: https://www.youtube.com/watch?v=cvTchBdrqdw around 4:30)
Is it same BRDF for curly vs straight that just visually interacting differently because of the shape of the hair itself?
AbhiAlderman
Would different types of hair require different algorithms to render correctly? It seems to me that different hair types could also be thought of as different materials, as the way light interacts with them would be different. Curly hair probably wouldn't reflect as much light as straight hair would, but would that require a different rendering sequence to get right? Similarly, the same hair type can often have various looks, such as when it is wet, dry, frizzy, oily, etc. Would you need different algorithms for each of these conditions as well?
KevinXu02
I think there are different algorithms to handle hair under different situations, and most of them are costly. For instance https://www.cs.columbia.edu/cg/liquidhair/
Taken as a whole, both the marschner model and the kajiya kay model are both robust methods for rendering hair and fur. The simplicity of the kajiya kay might be preferred for when performance is necessary such as in game. By proposing hair as a collection of cylinders, we can use an anisotropic specular reflection. On the other hand, Marschner Model is a slightly more involved method. It is based on how hair naturally scatters as may have multiple colors and textures.
Is there any difference when rendering hair that has been styled quite differently? By quite differently I mean braids vs straight hair. Or curly vs straight. And then curly in a bun vs straight in a bun? My first instinct is to think yes, especially since curly hair interacts with its environment differently to straight hair. And then how does one apply this to hair that is up in a bun? Are bun styles just a sphere with hair texture and BRDF? Or a braid is just a manipulated polygon with various volumes to represent each "chunk" of the braid? (referring to: https://www.youtube.com/watch?v=cvTchBdrqdw around 4:30) Is it same BRDF for curly vs straight that just visually interacting differently because of the shape of the hair itself?
Would different types of hair require different algorithms to render correctly? It seems to me that different hair types could also be thought of as different materials, as the way light interacts with them would be different. Curly hair probably wouldn't reflect as much light as straight hair would, but would that require a different rendering sequence to get right? Similarly, the same hair type can often have various looks, such as when it is wet, dry, frizzy, oily, etc. Would you need different algorithms for each of these conditions as well?
I think there are different algorithms to handle hair under different situations, and most of them are costly. For instance https://www.cs.columbia.edu/cg/liquidhair/