A few years ago I tried contacting the first author (who did this research as her PhD thesis) to ask for any code or algorithm calibration parameters they had.

A few years ago I tried contacting the first author (who did this research as her PhD thesis) to ask for any code or algorithm calibration parameters they had. I also contacted the university (to ask for the thesis) and some other people I found on the web who were listed as contributors or supervisors to this work (although that didn't include the other authors of this paper, maybe it wasn't published yet back then?) Never heard from any of them, even after several tries. Rather disappointing - usually people are jumping with joy that others have read their work and will jump at any chance to send material (I know I am). I don't know if they thought they could make a product out of it, but if they did or tried, I've never seen or found it. That said, what I found missing from the research was a comparison of this (highly computationally intensive) statistics-based line rendering with other, move naive (stochastical) but (much) faster line drawing algorithms. It's not hard to come up with a modification of Bresenham's algorithm that adds some (Perlin) noise to stroke width and color intensity (as a proxy for pressure), and which produces (in my subjective, anecdotal judgement) a lot like hand-drawn strokes. (I realize my first paragraph puts a moral imperative on me posting a link to my code of this, but alas my two evenings of work on this were wiped away when I thought it would be a good idea to experiment with using Scons as a build system and then accidentally switching the order of two arguments somewhere, causing my .cpp file to be overwritten with an empty file).