US Patent Issued to NVIDIA on Feb. 10 for "Performing visual relational reasoning" (California, Massachusetts, Washington, Texas Inventors)

ALEXANDRIA, Va., Feb. 11 -- United States Patent no. 12,547,893, issued on Feb. 10, was assigned to NVIDIA Corp. (Santa Clara, Calif.).

"Performing visual relational reasoning" was invented by Xiaojian Ma (Los Angeles), Weili Nie (Sunnyvale, Calif.), Zhiding Yu (Santa Clara, Calif.), Huaizu Jiang (Amherst, Mass.), Chaowei Xiao (Seattle), Yuke Zhu (Austin, Texas) and Anima Anandkumar (Pasadena, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A vision transformer (ViT) is a deep learning model that performs one or more vision processing tasks. ViTs may be modified to include a global task that clusters images with the same concept together to produce semantically consistent relational representations, as...