Eastern Ink Painting on a Computer

Roland Piquepaille writes "Traditional Oriental ink painting is more easily done with real brushes than with a computer program because you need to model how the ink is flowing into an absorbent surface such as paper. In this brief article, Technology Research News writes that "researchers from the Hong Kong University of Science and Technology have developed a brush-and-ink-style paint program, dubbed MoXi, that uses a model of pigment particles in water flowing into paper." These virtual Chinese brushes simulate in real time the ink dispersion and could be available on your PC within two years. This longer overview contains more details and references. It also includes pictures generated with MoXi. Finally, it looks at a potential trademark problem over the name MoXi."

Comments

...any bets?

[erroneus]

Any bets on how long it will take before this is emulated in a Photoshop or (hopefully) GiMP plugin? While I appreciate the achievment this represents, it's far from earth-shattering.

Still, the other part of the prize is the user interface. After all, the pressure of the brush plays an important role in all of that. I have seen programs where the speed of the brush movement helps to simulate the pressure of the brush but it's not natural enough.

I hope the need is important enough to justify the work done in this case.

Re:ads

[JanneM]

Why not just filter out the ads and go to the site?

And if he really does do a good writeup, then more power to him.

Input Devices

[Sideswiped]

I'm very excited to see the simulation of pigment on paper. I just hope that headway can be made on input devices. Wacom tablets are a nice start. Though the slickness of Wacom tablets can be a challenge to get used to. You can lay a sheet of paper over the tablet to add some resistance, but then the nib of the pen wear outs quickly.

Something in a the form of a force feedback pen would be amazing. One that you could set the resistance and the smooth/roughness of the surface you wish to emulate. If one already exists I would love to get some info.

Re:Input Devices

[sakusha]

There's the core issue. Big bucks were spent on researching and emulating an effect that can be done for pennies with real ink, paper, and a brush. Some things do not need to be simulated, they are so cheap and easy to do in the original media. The best input for brushwork is ink and paper, and a scanner. People often ask me how I get my brushwork effects in my computer graphics, I tell them I do brushwork with a real brush.

Cassidy Curtis's CG watercolors

[bleppie]

Sounds similar to the work Cassidy Curtis and folks at UW did on computer-generated watercolors using models of fluid flow: http://www.otherthings.com/uw/watercolor/

Ew

[Hikaru79]

Hm... personally, although the concept sounds absolutely *great* to me, the screenshots released so far are nothing to drool over, if you actually look at the article (and sell your soul to Roland). Download the high-res Lotus Leaves and take a look at the edges of those ink blots. No *way* that real ink would leave so clear and crisp an edge on absorbant, realistic paper as you can see there.

I realize that this is two years away from a release, but I really hope they introduce some major changes into their algorithm there, because so far they are not even competing for realism with products currently on the market...

If this doesn't sound like...

[FFFish]

Corel Natural Painter. It, too, simulates individual bristles, ink flow, pressure and angle, etcetera.

Interesting, but not new

[Dire Bonobo]

While interesting and pretty, this is not exactly cutting-edge research; the techniques for paint-and-water diffusion were laid out by Cassidy Curtis and his co-authors in 1997 (link), and instrumented haptic brushes with underlying simulated-brush models were examined in 2001 (brush models instrumented brush (pdf).

I suspect that's why people have been saying "hey, that seems a lot like [insert drawing program here]" - this is a (somewhat) new twist on old techniques. (Which, of course, is why it's a one-page submission to the conference, rather than a 10-pager like the original "Computer-Generated Watercolor". Not that there's anything wrong with that.)

thought struck me

[Anonymous Coward]

if you depiquepaille slashdot using your own fancy schmancy script, how is it that you're commenting on every primidi story?