US 11,657,230 B2
Referring image segmentation
Joon-Young Lee, San Jose, CA (US); and Seonguk Seo, Seoul (KR)
Assigned to ADOBE INC., San Jose, CA (US)
Filed by ADOBE INC., San Jose, CA (US)
Filed on Jun. 12, 2020, as Appl. No. 16/899,994.
Prior Publication US 2021/0390700 A1, Dec. 16, 2021
Int. Cl. G06T 7/10 (2017.01); G06F 40/30 (2020.01); G06F 40/20 (2020.01); G06F 16/9032 (2019.01); G06F 17/16 (2006.01); G06F 18/25 (2023.01); G06V 20/70 (2022.01)
CPC G06F 40/30 (2020.01) [G06F 16/90332 (2019.01); G06F 17/16 (2013.01); G06F 18/25 (2023.01); G06F 40/20 (2020.01); G06T 7/10 (2017.01); G06V 20/70 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for image processing, comprising:
receiving an input image and a referral expression corresponding to a target object in the input image;
generating an image mask corresponding to the target object based on an image feature vector representing the input image and a plurality of language feature vectors, wherein each of the language feature vectors comprises a representation of the referral expression in one of a plurality of language embedding spaces having a different number of dimensions from each other; and
displaying the target object by overlaying the image mask on the input image.