US 12,230,055 B2
Methods for creating personalized items using images associated with a subject and related systems and computers
Mike Vaggalis, Raleigh, NC (US); Erin Burchik, Lilburn, GA (US); John Petitte, Raleigh, NC (US); Paulo Rieck, Timbo (BR); Christina Dill, Ocilla, GA (US); and Paulo de Tarso Oliveira da Silva, Macapa (BR)
Assigned to Keepsake Tales Inc., Raleigh, NC (US)
Filed by Keepsake Tales Inc., Raleigh, NC (US)
Filed on Feb. 11, 2022, as Appl. No. 17/669,541.
Claims priority of provisional application 63/148,345, filed on Feb. 11, 2021.
Prior Publication US 2022/0254188 A1, Aug. 11, 2022
Int. Cl. G06K 9/00 (2022.01); G06T 11/60 (2006.01); G06V 10/20 (2022.01); G06V 40/10 (2022.01); G06V 40/16 (2022.01)
CPC G06V 40/165 (2022.01) [G06T 11/60 (2013.01); G06V 10/255 (2022.01); G06V 40/103 (2022.01); G06V 40/171 (2022.01); G06V 40/172 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A method for creating personalized items using images associated with a subject, the method comprising:
receiving and processing at least one image of the subject;
translating the processed at least one image into a personalized illustration having features that substantially resemble features of the subject, wherein translating the processed at least one image comprises extracting a plurality of core shapes from a face of the subject in the at least one image and assembling the extracted plurality of core shapes into a face of the personalized illustration that is completely different from the received at least one image; and
manipulating the personalized illustration of the subject for placement in a personalized item such that the manipulated personalized illustration appears as an integral part of an existing template of the personalized item,
wherein at least one of receiving, translating and manipulating are performed by at least one processor;
wherein translating comprises translating from the received at least one image to the personalized illustration using one of a vectorized approach and/or a machine learning approach; and
wherein the machine learning approach comprises recognizing different facial features in the received at least one image and matching the recognized different facial features to a correct permutation of pre-created facial assets to provide the personalized illustration.