The authors describe in mathematical rigor a computer simulation system for the stroke of a calligraphic brush. The examples provided are extremely accurate to the traditionally created analogues they attempt to render.

The key innovations of their simulation system, beyond the shape drift of the brush ellipse given changes in pressure over the course of a stroke are their simulation of ink uptake and release by the brush, and their ability to simulate the brush’s splitting under realistic pressure and ink level conditions.

They include a number of example images that they created with their tool, showing largely traditional chinese artwork and characters.

They reference
C.J. Curtis, S.E. Anderson, J.E. Seims, K.W. Fleischer, D.H. Salesin, Computer-generated
watercolor, Proc. SIGGRAPH 97 (1997) 421–430,
another very thorough system for simulation of watercolor, though it is far more paper/ink interaction based rather than brush focused.

The interfaces for all these systems are very software focused; the control is not manual but rather on cumbersome software settings; ultimately the mouse and keyboard are all that are used to control the movement of the brushes.

This indirect control enforces a rather extreme time commitment for very basic results; there is very little motivation to use the tool over its physical inspiration.

The cohesion of the various fibers of the calligraphic brush given purpose through ink is a rich material phenomenon that parallels the unity ideal that forms a strong thread within the oriental aesthetic. It motivates tracing the evolution of calligraphy over time.

Canonically, though, this software is more enabling to the automation of calligraphy-like imagery than it is to the continuation of traditional calligraphic forms into the digital age. What computer data might we transliterate into this newly manipulable medium? What kind of information begs representation in the fractured but coherent strokes of a calligraphic pen?

The authors use a 6-dimensional simulation state machine for the brush: are all of these dimensions transparently encoded in the output? Are they responsible for our attribution of complexity to the brush strokes?

Songhua Xua, b, Min Tanga, Francis C. M. Lau, and Yunhe Pana. Virtual hairy brush for painterly rendering, sciencedirect.com 2004.