Systems and methods for classifying and representing gestural inputs
AssigneeAT&T Intellectual Property II, L.P.
InventorsMichael J. Johnston, Srinivas Bangalore
Life of this patent
Estimating the term…
Abstract
Gesture and handwriting recognition agents provide possible interpretations of electronic ink. Recognition is performed on both individual strokes and combinations of strokes in the input ink lattice. The interpretations of electronic ink are classified and encoded as symbol complexes where symbols convey specific attributes of the contents of the stroke. The use of symbol complexes to represent strokes in the input ink lattice facilitates reference to sets of entities of a specific type.
Claims36
A method for representing gestures made by a user, comprising:
- recognizing a gesture;
- converting the recognized gesture into at least one recognition result by using finite-state transducers, each recognition result comprising a sequence of symbols including a set of gesture symbols with a SEM (semantics) placeholder symbol for specific content of the gesture, wherein each symbol of the sequence of symbols identifies a specific attribute of the recognized gesture and the SEM placeholder symbol represents the specific content of the gesture such that when recomposing specific gestural content, the specific content is mapped to the SEM placeholder symbol and other specific attributes are mapped to themselves;
- and
- outputting the recognition result.
A system for recognizing strokes made on an input device, comprising:
- a recognition device usable to recognize gestures;
- and
- a converting device that converts the recognized gesture into at least one recognition result by using finite-state transducers, each recognition result comprising a sequence of symbols including a set of gesture symbols with a SEM (semantics) placeholder symbol for specific content of the gesture, wherein each symbol of the sequence of symbols identifies a specific attribute of the recognized gesture and the SEM placeholder symbol represents the specific content of the gesture such that when recomposing specific gestural content, the specific content is mapped to the SEM placeholder symbol and other specific attributes are mapped to themselves;
- and
- an output device that outputs the recognition result.
A method for free-form gesture representation, comprising:
- recognizing a gesture;
- converting the recognized gesture into at least one recognition result by using finite-state transducers, each recognition result comprising a sequence of symbols including a set of gesture symbols with a SEM (semantics) placeholder symbol for specific content of the gesture, wherein each symbol of the sequence of symbols identifies a specific attribute of the recognized gesture and the SEM placeholder symbol represents the specific content of the gesture such that when recomposing specific gestural content, the specific content is mapped to the SEM placeholder symbol and other specific attributes are mapped to themselves;
- generating a recognition lattice based on the recognition result;
- and
- outputting the recognition lattice.
Dependencies are inferred from the claim text.