Today I’ll be discussing two principles of visual design and how they are utilized to create effective characters. The two principles can be thought of as two ends sides of a spectrum: Abstract and Structural.
Abstracted characters are often times considered very cartoon-y and simple. Because it is barely tethered by the rules of physicality, it allows for the most visual freedom. The basic building blocks of character creation for abstract characters start with shapes. To fully master abstract character design, it requires knowledge of iconography. Frequently, abstract designs have the most iconic silhouettes, however it can also carry a cultural burden of being “childish”.
Examples of mostly abstracted characters: Mickey Mouse, Pac-Man, chibi anime, etc.
Structural characters are based in physicality and tangibility. Most people would consider it “realistic” …which implies that the intangible isn’t real, BUT I DIGRESS. To utilize this form of character building, it requires a knowledge of physicality. The more structural a character, the more likely it is that photos, real life costuming/objects, and 3D models will be incorporated in some way.
Examples of mostly structural characters: Pop stars, live-action characters/creatures (no CGI), photo-realistic characters/creatures.
These principles have a lot of similarities to flat and skeuomorphic interface design, but unlike interface design, character design involves mixing and blending the two principles together to a higher degree. It’s also incredibly rare to have any character be 100% abstract or 100% structural. I can’t list every possible way a design might mix abstract and structural elements so I’ll list off three common techniques:
1) Start with an abstract base, work towards structural. Often used to create interesting silhouettes using abstracted shapes. Structure is then added in through texture, detail, and mechanics.
A great example of this is illustrated in the development of Rango characters. On the left is a heavily abstracted concept by Eugene Yelchin that uses interesting shapes with little to no structural integrity. Then in the middle, Crash McCreery’s final 2D design with a layer of structure added with gravity, texture, and color. Then finally, we have the 3D model that is fully volumetric, boned, rigged, lit, and textured.
Another example that clearly illustrates the concept is R2D2. The main draw of his design is his round head and stumpy, non-threatening silhouette. The buttons and mechanisms on him, though still important, are secondary. In the next section I’ll bring up another mechanical character as a comparison.
To a lesser degree, we can see the same thing in the line up of characters from The Triplets of Belleville. The characters’ bodies are warped and stretched to create interesting shapes. However, despite how impossibly exaggerated their physiques are, there’s enough structure added in for us to imagine a working bone and muscular system inside of them (albeit heavily distorted).
2) Start with a structural base, work towards abstract. Take an existing base character rooted in functionality/workable parts, and accentuate only the most desired parts of the character.
One example is the Pale Man who blindly eats children being both underfed and overfed at the same time. See how his costume and the green screens achieved that look by artificially exaggerating his thinness, rearranging his eyes, and adding excess flesh.
Also, remember R2D2’s design? Here’s another mechanical character to compare and contrast:
Both R2D2 and the Terminator are robots who have to look believable next to real actors, but unlike, R2D2, whose mechanical functionality runs off the imagination of the audience, it wouldn’t be surprising if the Terminator’s parts were created by engineers to be moderately workable and then tweaked to be fully function.
3) You can isolate entire parts of a character to be abstracted or structural. The most common part of a character that gets isolated is probably the face, which is what I will be focusing on for now.
Notice in the image above, how the level of fidelity is simplified around the face to allow for greater emotional expression, while the rest of Korra is more structural. That’s not to say that her body isn’t also simplified, but instead that it respects bone and muscle structure much more strictly than her face does.
Another example of this is seen when working in real life where storytellers use masks to abstract faces.
See also how artists create personifications of youth and beauty by accentuating traits that are pretty much exclusive to young people: Small noses and ears (which never stop growing as you age), slim waists, healthy hair, and a fresh complexion untouched by life’s hardships.
Let’s also not forget, faces can get the opposite treatment and have structure added in… though usually for a jarring effect.
Now I just want to take this time to reiterate that there are a billion different ways to land on the abstract vs. structural spectrum. These techniques aren’t meant to be strictly abided by, but instead used as a starting point to expand upon. With that said, I’ll end with Michael Paulus’s cartoon studies, where he super imposes a skeletal structure into highly abstracted cartoon characters.
Thanks for reading!