US 12,475,680 B2
Method of training image representation model
Minyoung Mun, Suwon-si (KR); and Seijoon Kim, Suwon-si (KR)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Mar. 2, 2023, as Appl. No. 18/116,602.
Claims priority of application No. 10-2022-0111092 (KR), filed on Sep. 2, 2022.
Prior Publication US 2024/0078785 A1, Mar. 7, 2024
Int. Cl. G06V 10/764 (2022.01); G06V 10/74 (2022.01)
CPC G06V 10/761 (2022.01) [G06V 10/764 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A training method performed by a computing apparatus, the training method comprising:
generating an anchor image embedding vector for an anchor image using an image representation model;
determining first similarities between the anchor image and negative samples of the anchor image using first image embedding vectors for the negative samples and the generated anchor image embedding vector;
determining second similarities between the anchor image and positive samples of the anchor image using second image embedding vectors for the positive samples and the generated anchor image embedding vector;
obtaining one of a vector corresponding to a label of the anchor image and third similarities between the label of the anchor image and labels of the negative samples;
determining a loss value for the anchor image based on (i) the determined first similarities, (ii) the determined second similarities, and (iii) one of the obtained third similarities and a fourth similarity, wherein the fourth similarity is a similarity between the obtained vector and the generated anchor image embedding vector; and
updating weights of the image representation model based on the determined loss value.