US 11,836,187 B2
Generating occurrence contexts for objects in digital content collections
Manoj Kilaru, Hyderabad (IN); Vishwa Vinay, Bangalore (IN); Vidit Jain, Delhi (IN); Shaurya Goel, Hisar (IN); Ryan A. Rossi, Santa Clara, CA (US); Pratyush Garg, New Delhi (IN); Nedim Lipka, Campbell, CA (US); and Harkanwar Singh, Patiala (IN)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Oct. 26, 2020, as Appl. No. 17/079,945.
Prior Publication US 2022/0129498 A1, Apr. 28, 2022
Int. Cl. G06F 16/55 (2019.01); G06N 20/00 (2019.01); G06F 16/583 (2019.01); G06F 16/54 (2019.01); G06F 40/47 (2020.01); G06F 40/30 (2020.01); G06F 18/214 (2023.01)
CPC G06F 16/5854 (2019.01) [G06F 16/54 (2019.01); G06F 16/55 (2019.01); G06F 18/214 (2023.01); G06F 40/30 (2020.01); G06F 40/47 (2020.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. In a digital medium environment to determine contexts in which objects appear in a digital image collection, a method implemented by a computing device, the method comprising:
receiving, by a representation learning module, context request data describing an object that is depicted with additional objects in digital images of the digital image collection;
generating, by the representation learning module, relationship embeddings for the object and each of the additional objects using a representation learning model trained to predict relationships for objects;
forming, by a graph module, a relationship graph for the object that includes a vertex for each relationship between the object and the additional objects indicated by the relationship embeddings;
clustering, by a cluster module, vertices of the relationship graph into contextual clusters that each represent an occurrence context of the object in the digital images of the digital image collection; and
generating, by the cluster module and for each contextual cluster, an indication of a respective occurrence context for the object for display in a user interface of a display device.