US2023234235A1PendingUtilityA1

Method for generating a handwriting vector

Assignee: CONV INCPriority: Sep 18, 2019Filed: Jan 23, 2023Published: Jul 27, 2023
Est. expirySep 18, 2039(~13.2 yrs left)· nominal 20-yr term from priority
G06T 11/23B25J 11/00G06T 11/60B25J 11/0005G06F 40/109G06V 30/347G06V 30/413G06V 10/82G06V 10/764G06V 10/44G06F 18/2413
66
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

One variation of a method includes: accessing a handwriting sample comprising a set of user glyphs handwritten by a user; for each character in a set of characters, identifying a subset of user glyphs corresponding to the character in the handwriting sample, characterizing variability of a set of spatial features across the subset of user glyphs, and storing variability of the set of spatial features across the subset of user glyphs in a character container corresponding to the character; and compiling the set of character containers into a handwriting model for the user. The method further includes: accessing a text string comprising a combination of characters in the set of characters; for each instance of each character in the text string, inserting a set of variability parameters into the handwriting model to generate a synthetic glyph representing the character; and assembling the set of synthetic glyphs into a print file.

Claims

exact text as granted — not AI-modified
I claim: 
     
         1 . A method comprising:
 during a first time period:
 accessing a handwriting sample comprising a set of user glyphs handwritten by a user and representing a set of characters; 
 for each character in the set of characters:
 identifying a subset of user glyphs, in the handwriting sample, corresponding to the character; 
 characterizing a set of spatial features representative of the character across the subset of user glyphs; and 
 storing the set of spatial features in a character container, in a set of character containers, corresponding to the character; and 
 
 compiling the set of character containers into a handwriting model associated with the user; and 
   during a second time period succeeding the first time period:
 accessing a text string comprising a combination of characters; 
 for each instance of each character in the text string, generating a synthetic glyph, in a set of synthetic glyphs, representing the character based on the handwriting model; and 
 assembling the set of synthetic glyphs into a print file. 
   
     
     
         2 . A method comprising:
 accessing a handwriting sample of a user;   identifying a first set of user glyphs representative of a first character, in a set of characters, represented in the handwriting sample;   for a first user glyph, in the first set of user glyphs, corresponding to a first instance of the first character, extracting a first set of spatial features representative of the first character;   for a second user glyph, in the first set of user glyphs, corresponding to a second instance of the first character, extracting a second set of spatial features representative of the first character and analogous the first set of spatial features;   interpolating a first set of synthetic spatial features based on the first set of spatial features of the first user glyph and the second set of spatial features of the second user glyph;   generating a first synthetic glyph of the first character based on the first set of synthetic spatial features;   compiling the first user glyph, the second user glyph, and the first synthetic glyph into a first character container, in a set of character containers, corresponding to the first character; and   compiling the set of character containers into a handwriting model associated with the user.   
     
     
         3 . A method comprising:
 during a first time period:
 accessing a handwriting sample comprising a set of user glyphs handwritten by a user and representing a set of characters; 
 for each character in the set of characters:
 detecting a subset of user glyphs, in the set of user glyphs, corresponding to the character; 
 identifying a set of spatial features representative of the character across the subset of user glyphs; 
 initializing a set of coefficients within a character container, in a set of character containers, corresponding to the character based on the set of spatial features of the subset of user glyphs; 
 initializing a variability vector; 
 inserting the set of coefficients and the variability vector into a synthetic handwriting generator to generate a synthetic glyph of the character; 
 characterizing a difference between the synthetic glyph and the subset of user glyphs corresponding to the character; and 
 in response to the difference exceeding a threshold difference, adjusting the set of coefficients and the variability vector to reduce the difference; and 
 
 compiling the set of character containers into a handwriting model associated with the user; and 
   during a second time period succeeding the first time period:
 accessing a text string comprising a combination of characters in the set of characters; 
 for each instance of each character in the text string, generating a synthetic glyph, in a set of synthetic glyphs, representing the character based on the handwriting model; and 
 assembling the set of synthetic glyphs into a print file.

Join the waitlist — get patent alerts

Track US2023234235A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.