US 12,243,308 B2
Learning iconic scenes and places with privacy
Michael Chatzidakis, San Jose, CA (US); Kalu O. Kalu, Los Gatos, CA (US); Omid Javidbakht, Cupertino, CA (US); Sowmya Gopalan, Cupertino, CA (US); Eric Circlaeys, Los Gatos, CA (US); Rehan Rishi, San Jose, CA (US); and Mayank Yadav, Paris (FR)
Assigned to Apple Inc., Cupertino, CA (US)
Filed by Apple Inc., Cupertino, CA (US)
Filed on Apr. 8, 2022, as Appl. No. 17/658,474.
Claims priority of provisional application 63/197,252, filed on Jun. 4, 2021.
Prior Publication US 2022/0392219 A1, Dec. 8, 2022
Int. Cl. G06F 16/90 (2019.01); G06F 16/906 (2019.01); G06F 16/908 (2019.01); G06V 20/50 (2022.01)
CPC G06V 20/50 (2022.01) [G06F 16/906 (2019.01); G06F 16/908 (2019.01)] 17 Claims
OG exemplary drawing
 
1. A device, comprising:
a memory; and
one or more processors operatively coupled to the memory, wherein the one or more processors are configured to execute instructions causing the one or more processors to:
identify a first set of digital assets (DAs) related to a user of the device, wherein each digital asset in the first set of digital assets matches each of one or more specified criteria received from a server device, and wherein each of the one or more specified criteria correspond to at least one property of a digital asset, wherein a first property corresponding to at least one of the one or more specified criteria comprises a scene content property of a digital asset, and wherein a second property corresponding to at least one of the one or more specified criteria comprises a capture location property of a digital asset;
generate a first list of image property sets,
wherein the first list comprises an image property set for each digital asset in the first set of digital assets, and
wherein each image property set in the first list comprises, for a particular digital asset from the first set of digital assets:
a group of values for each of the particular digital asset's properties corresponding to the one or more specified criteria;
generate a second list of image property sets,
wherein the second list comprises a first number of noise-injecting image property sets, and
wherein each noise-injecting image property set in the second list comprises:
a group of generated digital asset property values corresponding to the one or more specified criteria, wherein at least one of the generated digital asset property values comprises a randomly-generated digital asset property value, and wherein each noise-injecting image property set in the second list further comprises at least: a first randomly-generated digital asset property value for the first property; and a second randomly-generated digital asset property value for the second property; and
submit, to the server device, a third list of image property sets, wherein the third list of image property sets comprises the first list of image property sets and the second list of image property sets.