In this paper we propose an approach for constructing sketch parsers whose recognition accuracy and speed is significantly improved by acquiring information on the user?s sketching style during a training phase. The construction process consists in specifying a sketch grammar description of the language syntax, automatically generating a parser from such specification, and let the user train the recognition system on a set of sketch sentences.
Citation:
Gennaro Costagliola, Vincenzo Deufemia, Michele Risi, "A Trainable System for Recognizing Diagrammatic Sketch Languages," vlhcc, pp.281-283, 2005 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC'05), 2005