ALEXANDRIA, Va., June 16 -- United States Patent no. 12,307,694, issued on May 20, was assigned to Toyota Research Institute Inc. (Los Altos, Calif.).
"Self-supervised monocular depth estimation via rigid-motion embeddings" was invented by Rares A. Ambrus (San Francisco), Sergey Zakharov (San Francisco), Vitor Guizilini (Santa Clara, Calif.) and Adrien David Gaidon (Mountain View, Calif.).
According to the abstract* released by the U.S. Patent & Trademark Office: "System, methods, and other embodiments described herein relate to training a depth model for monocular depth estimation using photometric loss masks derived from motion estimates of dynamic objects. In one embodiment, a method includes generating depth maps from images of an env...