US 12,169,936 B2
Semi supervised training from coarse labels of image segmentation
Abhishek Chaurasia, Bellevue, WA (US); and Katya B. Giannios, West Lafayette, IN (US)
Assigned to Micron Technology, Inc., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on Apr. 22, 2022, as Appl. No. 17/727,604.
Claims priority of provisional application 63/185,278, filed on May 6, 2021.
Prior Publication US 2022/0358658 A1, Nov. 10, 2022
Int. Cl. G06T 7/10 (2017.01); G06N 3/045 (2023.01)
CPC G06T 7/10 (2017.01) [G06N 3/045 (2023.01); G06T 2200/24 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, in a computing apparatus, first data representative of first images and second data identifying image segments in the first images;
receiving, in the computing apparatus, third data representative of second images and fourth data identifying approximate image segments in the second images;
training, by the computing apparatus, a first artificial neural network to perform image segmentation on the first images and the second images according to the image segments identified in the second data and the approximate image segments identified in the fourth data;
performing, by the computing apparatus using the first artificial neural network, image segmentation of the second images to generate fifth data identifying image segments in the second images; and
training, by the computing apparatus, a second artificial neural network to perform image segmentation on the first images and the second images according to the image segments identified in the second data and the image segments identified in the fifth data.