US 12,277,633 B2
Create editable vectorized image from bitmap image
Markiyan Kostiv, Lviv (UA); Sam Raven Oliver Eckert, Stuttgart (DE); Igor Filipe Viveiros de Assis, Berlin (DE); Ralph Theodori, Berlin (DE); Oleksandra Boiovych, Odesa (UA); and Mihai Vladimir Danila, Hilden (DE)
Assigned to LINEARITY GMBH, Berlin (DE)
Filed by Linearity GmbH, Berlin (DE)
Filed on Nov. 14, 2022, as Appl. No. 17/986,157.
Prior Publication US 2024/0161363 A1, May 16, 2024
Int. Cl. G06T 11/60 (2006.01); G06T 3/40 (2024.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01)
CPC G06T 11/60 (2013.01) [G06T 3/40 (2013.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01); G06T 2200/24 (2013.01); G06T 2210/36 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
importing an image;
classifying the image;
selecting a first algorithm from a plurality of algorithms, based on an image classification;
providing a visual indication, on a graphical user interface (GUI), of the first algorithm;
receiving an input from a user, via the GUI, indicating that the user selects a second algorithm for vectorizing the image, wherein, upon classifying the image as a photograph, the second algorithm comprises:
performing statistical region merging to group neighboring pixels in the image based on a similarity of their color shades within a threshold;
splitting a set of colors of the image into a set of buckets and merging the set of buckets into regions; and
identifying boundaries of the regions and establishing outline paths;
vectorizing the image using the second algorithm; and
providing an editable version of the vectorized image.