CALI: An Online Scribble Recognizer for Calligraphic Interfaces

Abstract

CALI is a fast, simple and compact online recognizer that identifies Scribbles (multi-stroke geometric shapes) drawn with a stylus on a digitizing tablet. Our method is able to identify shapes of different sizes and rotated at arbitrary an-gles, drawn with dashed, continuous strokes or overlapping lines. We use temporal adjacency to allow users to input the most common shapes in drawing such as triangles, lines, rect-angles, circles, diamonds and ellipses, using multiple strokes. We have further extended this approach to identify useful shapes such as arrows, crossing lines and unistroke gesture commands and have developed a library of software compo-nents to make this software generally available. The recog-nition algorithm uses Fuzzy Logic and geometric features, combined with an extensible set of heuristics to classify scrib-bles. More recently we developed a trainable version of the recognizer to allow users to easily add new shape classes to the initial core set. Evaluation results show recognition rates over 97 % for the non-trainable and 95 % for the trainable ver-sion.