US Patent Issued to SAMSUNG SDS AMERICA on Dec. 9 for "Bayesian semantic segmentation active learning with beta approximation" (California Inventors)

ALEXANDRIA, Va., Dec. 9 -- United States Patent no. 12,494,046, issued on Dec. 9, was assigned to SAMSUNG SDS AMERICA INC. (San Jose, Calif.).

"Bayesian semantic segmentation active learning with beta approximation" was invented by Sima Didari (San Jose, Calif.), Jae Oh Woo (Fremont, Calif.), Heng Hao (San Jose, Calif.), Hankyu Moon (San Ramon, Calif.) and Patrick Bangert (Sunnyvale, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Training of a machine vision model, a segmentation model, is performed by using an acquisition function for a small number of pixels of one or more training images. The acquisition function uses first mutual information and second mutual information to identify unlabelled pix...