Reverse handwriting: from ink to word

Abstract

We present a principled approach for cursive handwriting recognition which builds upon handwriting generation models. According to such models handwriting is a learned complex motoric task which is accomplished by sequencing simpler movement called stroke. As learning proceeds in human, so does fluency, which results in producing similar sequence of strokes in correspondence of the same sequence of letters. Such invariants represents therefore the basic drawing units to which an interpretation can be associated. Recognition is then achieved by detecting the invariants used to produce the word to be recognized, associating to them their interpretations, and eventually concatenating the interpretations along the ink of the word. Experiments on on-line data of the current implementation are reported and discussed to show the effectiveness of the method.