Freeform digital ink annotation recognition
First Claim
1. A computer-implemented system that facilitates recognition, comprising:
- at least one processor configured to execute the following computer-executable components;
a component that receives at least one input containing digital ink strokes;
a computer-executable recognition component that identifies and groups at least a subset of the digital ink strokes from the input into at least one recognized annotation;
a computer-executable classification component executing from a computer memory employing at least one type detector that classifies the recognized annotation into at least one type of a group of extensible types, the at least one type detector determines a best idealized version of the subset of the digital ink strokes of the at least one recognized annotation determined according to the at least one type using stroke features that capture similarity of the subset of the digital ink strokes with an idealized version of an annotation associated with the at least one type detector;
a resolution component that facilitates the recognition of the subset of the digital ink strokes by maximizing a number of explained strokes, maximizing an overall confidence, and minimizing a number of hypotheses for the input, the resolution component optimizes recognition of the annotation via employment of an energy function given by;
where α and
β
are empirically-determined weights.
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention leverages classification type detectors and/or context information to provide a systematic means to recognize and anchor annotation strokes, providing reflowable digital annotations. This allows annotations in digital documents to be archived, shared, searched, and easily manipulated. In one instance of the present invention, an annotation recognition method obtains an input of strokes that are grouped, classified, and anchored to underlying text and/or points in a document. Additional instances of the present invention utilize linguistic content, domain specific information, anchor context, and document context to facilitate in correctly recognizing an annotation.
112 Citations
41 Claims
-
1. A computer-implemented system that facilitates recognition, comprising:
-
at least one processor configured to execute the following computer-executable components; a component that receives at least one input containing digital ink strokes; a computer-executable recognition component that identifies and groups at least a subset of the digital ink strokes from the input into at least one recognized annotation; a computer-executable classification component executing from a computer memory employing at least one type detector that classifies the recognized annotation into at least one type of a group of extensible types, the at least one type detector determines a best idealized version of the subset of the digital ink strokes of the at least one recognized annotation determined according to the at least one type using stroke features that capture similarity of the subset of the digital ink strokes with an idealized version of an annotation associated with the at least one type detector;
a resolution component that facilitates the recognition of the subset of the digital ink strokes by maximizing a number of explained strokes, maximizing an overall confidence, and minimizing a number of hypotheses for the input, the resolution component optimizes recognition of the annotation via employment of an energy function given by;where α and
β
are empirically-determined weights.- View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A method for facilitating computer-implemented recognition, comprising:
-
employing a processor to execute computer executable instructions stored in memory to perform the following acts; receiving at least one input containing digital ink strokes; identifying and grouping at least a subset of the digital ink strokes from the input into at least one recognized annotation; classifying the recognized annotation into at least one type of a group of extensible types based in part on the output of at least one computer-executable type detector; determining with the at least one computer-executable type detector a best idealized version of the subset of the digital ink strokes of the at least one recognized annotation, the best idealized version is determined according to the at least one type using stroke features that determine similarity of the subset of the digital ink strokes of the at least one recognized annotation to an idealized version of an annotation associated with the at least one computer executable type detector; recognizing the annotation associated with the subset of the digital ink strokes by maximizing a number of explained strokes, maximizing an overall confidence, and minimizing a number of hypotheses for the input, wherein the recognition of the annotation is optimized via employment of an energy function given by; where α and
β
are empirically-determined weights.- View Dependent Claims (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40)
-
-
41. A system that facilitates recognition, comprising:
-
means for receiving at least one input containing digital ink strokes; means for storing the received input; means for identifying and grouping at least a subset of the digital ink strokes from the input into at least one recognized annotation by maximizing a number of explained strokes, maximizing an overall confidence, and minimizing a number of hypotheses for the input, wherein the recognition of the annotation is optimized via employment of an energy function given by; where α and
β
are empirically-determined weights;means for classifying the recognized annotation into at least one type of a group of extensible types based in part on the output of at least one type detector; means for determining a best idealized version of the at least a subset of the digital ink strokes of the at least one recognized annotation according to the at least one type using stroke features of the at least a subset of the digital ink strokes that facilitate in determining similarity of the best idealized version to an idealized version of an annotation associated with the at least one type detector; and means for comparing the fit of the best idealized version of the at least a subset of the digital ink strokes with contextual features associated with the at least one input.
-
Specification