US 11,971,885 B2
Retrieval aware embedding
Fengbin Chen, San Jose, CA (US); Venkat Barakam, Cupertino, CA (US); Benjamin Leviant, Mountain View, CA (US); Amine Ben Khalifa, Sunnyvale, CA (US); Kerem Turgutlu, Ames, IA (US); Jayant Kumar, San Jose, CA (US); Sumeet Zaverilal Gala, Milpitas, CA (US); Gaurav Kukal, Fremont, CA (US); and Vipul Dalal, Cupertino, CA (US)
Assigned to ADOBE INC., San Jose, CA (US)
Filed by ADOBE INC., San Jose, CA (US)
Filed on Feb. 10, 2021, as Appl. No. 17/172,986.
Prior Publication US 2022/0253435 A1, Aug. 11, 2022
Int. Cl. G06F 16/00 (2019.01); G06F 16/245 (2019.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01)
CPC G06F 16/245 (2019.01) [G06N 3/04 (2013.01); G06N 3/08 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for information retrieval, comprising:
generating a dense image embedding for each of a plurality of media objects to be searched;
generating a sparse image embedding with a higher number of dimensions than the dense image embedding for each of the media objects using an encoder that takes the dense image embedding as an input, wherein the sparse image embedding is generated using an activation function including a sparsity constraint that limits a number of non-zero parameters at a hidden layer of the encoder;
performing a search on the plurality of media objects based at least in part on the sparse image embedding; and
retrieving a matching object from the plurality of media objects based on the search.