Diffusion Curves

Stumbled upon an amazing SIGGRAPH paper on “Diffusion Curves: A Vector Representation for Smooth-Shaded Images”, at the Adobe ATL page : link. Given cartoon sketches represented as Bezier curves, with initial color values on either sides of the curves, they construct an image by letting the colors diffuse as governed by a Poisson equation. This can be used to rapidly create cartoon images and animations. The paper also solves the inverse problem, i.e. given an arbitrary smoothly shaded image, they determine the initial curves and color samples needed to vectorize the image.

Check out their cool demo here.

Diffusion Curves: A Vector Representation for Smooth-Shaded Images

Holger Winnemöller , Adobe Systems
Adrien Bousseau, INRIA, Grenoble, France
Alexandrina Orzan, INRIA, Grenoble, France
Pascal Barla, INRIA, Grenoble, France
Joelle Thollot, INRIA, Grenoble, France
David Salesin, Adobe Systems

3D morphable model face animation by Volker Blantz

Noise analysis with 2D time-freq histogram

First a tidbit:

” Humans have 200 million light receptors in their eyes, 10 to 20 million receptors devoted to smell, but only 8,000 dedicated to sound. Yet despite this miniscule number, the auditory system is the fastest of the five senses. Researchers credit this discrepancy to a series of lightning-fast calculations in the brain that translate minimal input into maximal understanding. And whatever those calculations are, they’re far more precise than any sound-analysis program that exists today.”
…
“Magnasco collaborated with Timothy Gardner, a former Rockefeller graduate student who is now a Burroughs Wellcome Fund fellow at MIT, to figure out how to get computers to process complex, rapidly changing sounds the same way the brain does. They struck upon a mathematical method that reassigned a sound’s rate and frequency data into a set of points that they could make into a histogram — a visual, two-dimensional map of how a sound’s individual frequencies move in time. When they tested their technique against other sound-analysis programs, they found that it gave them a much greater ability to tease out the sound they were interested in from the noise that surrounded it.

One fundamental observation enabled this vast improvement: They were able to visualize the areas in which there was no sound at all. The two researchers used white noise — hissing similar to what you might hear on an un-tuned FM radio — because it’s the most complex sound available, with exactly the same amount of energy at all frequency levels. When they plugged their algorithm into a computer, it reassigned each tone and plotted the data points on a graph in which the x-axis was time and the y-axis was frequency. “

– ( more ) —