US 12,242,570 B2
Method and apparatus for training image recognition model, and image recognition method and apparatus
Hong Shang, Shenzhen (CN); Han Zheng, Shenzhen (CN); and Zhongqian Sun, Shenzhen (CN)
Assigned to TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED, Shenzhen (CN)
Filed by TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED, Shenzhen (CN)
Filed on Feb. 12, 2024, as Appl. No. 18/438,595.
Application 18/438,595 is a continuation of application No. 17/515,312, filed on Oct. 29, 2021, granted, now 11,960,571.
Application 17/515,312 is a continuation of application No. PCT/CN2020/116998, filed on Sep. 23, 2020.
Claims priority of application No. 201910989262.8 (CN), filed on Oct. 17, 2019.
Prior Publication US 2024/0184854 A1, Jun. 6, 2024
Int. Cl. G06K 9/00 (2022.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06N 7/01 (2023.01); G16H 30/40 (2018.01); G16H 50/20 (2018.01)
CPC G06F 18/2155 (2023.01) [G06F 18/217 (2023.01); G06N 7/01 (2023.01); G16H 30/40 (2018.01); G16H 50/20 (2018.01); G06V 2201/03 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method for training an image recognition model, performed by a first electronic device, the method comprising:
obtaining training image sets, the training image sets comprising at least a first image set containing at least one first image, a second image set containing at least one second image and at least one perturbed image, and a third image set containing at least one third image, the first image being a labeled image corresponding to a first task, the second image being an unlabeled image corresponding to the first task, the third image being a labeled image corresponding to a second task, the first task and the second task being different tasks;
obtaining, by using an initial image recognition model, a first predicted probability outputted based on the first image set, a second predicted probability and a third predicted probability outputted based on the second image set, and a fourth predicted probability outputted based on the third image set;
determining a target loss function according to the first predicted probability, the second predicted probability, the third predicted probability, and the fourth predicted probability; and
training the initial image recognition model based on the target loss function, to obtain an image recognition model.