I’ve shaken the cobwebs off my real-time watercolor simulation. I’m reaching for something I’ve wanted to do for years: the technique known as lost and found edges. Here are a few baby steps in that direction.

My first naive implementation is actually a simple, two-step version of what we described in our 1997 paper: a series of wet-in-wet glazes, first wetting a large region, and then applying pigment (and more water) to a smaller region within it. Depending on the shapes of the regions, the paint flows in different ways. If you’re lucky, you end up with pigment that reaches the edge of the water in some places but not others.

More to say about this later, but for now, it feels great to be back at it.