This paper presents a smart interface that automatically extracts and rnes
pen strokes from images of hand
drawn sketches. The interface allows users to digitize hand-drawn material
such sketches of wcharts, cartoons
or other pen based drawings and automatically isolate and rne the
individual strokes making up the sketch. First,
we present a method for extracting pen strokes based on learned constraints
on curves. The approach consists of
using a training set that shows good examples of curves and how a user
would draw them. Given an image of
a hand-drawn sketch, the system selects the pen stroke that is most
statistically consistent with the examples in
the training set and ranks the others based on their likelihood. Users can
keep the selected candidate or they
may scroll through the other top candidates to select a preferred
solution. Second, we present an overview of our
rnement procedure and its application on the extracted pen strokes. Using
the same database of examples, the
extracted pen stroke is rned to make it 'look' more like those in the
database.