US 12,293,523 B2
Systems and methods for automatically cropping digital images
Sanchit Sanchit, Vienna (AT); and Paul Angerer, Vienna (AT)
Assigned to Canva Pty Ltd, Surry Hills (AU)
Filed by Canva Pty Ltd, Surry Hills (AU)
Filed on Mar. 8, 2024, as Appl. No. 18/600,625.
Claims priority of application No. 2023202007 (AU), filed on Mar. 31, 2023.
Prior Publication US 2024/0331160 A1, Oct. 3, 2024
Int. Cl. G06T 7/00 (2017.01); G06T 3/40 (2006.01); G06T 3/60 (2006.01); G06T 7/11 (2017.01); G06V 10/25 (2022.01); G06V 10/46 (2022.01)
CPC G06T 7/11 (2017.01) [G06T 3/40 (2013.01); G06T 3/60 (2013.01); G06V 10/25 (2022.01); G06V 10/462 (2022.01); G06T 2207/20132 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer implemented method for automatically generating a cropped version of an input image, the method including:
processing the input image to detect a salient region of the input image;
generating a set of crop candidates based on the salient region of the input image, wherein each crop candidate in the set of crop candidates defines a different image crop and generating the set of crop candidates includes:
determining a first target aspect ratio;
determining a first seed point, wherein the first seed point corresponds to a centre of the salient region;
determining a second seed point, wherein the second seed point is offset from the centre of the salient region;
generating a first subset of crop candidates based on the first seed point and the first target aspect ratio; and
generating a second subset of crop candidates based on the second seed point and the first target aspect ratio;
calculating a set of candidate scores, each candidate score corresponding to a crop candidate in the set of crop candidates;
selecting a first crop candidate from the set of crop candidates based on the set of crop candidate scores; and
generating the cropped version of the input image based on the image crop defined by the first crop candidate.