ALEXANDRIA, Va., April 21 -- United States Patent no. 12,608,792, issued on April 21, was assigned to UVeye Ltd. (Tel Aviv, Israel).
"Vehicle damage validation using symmetry-based deep learning" was invented by Shirel Gazit (London), Itai Orr (Or Akiva, Israel) and Amir Hever (Tenafly, N.J.).
According to the abstract* released by the U.S. Patent & Trademark Office: "A computer-implemented method and system for validating vehicle damage detection utilizes symmetry-based analysis of opposing vehicle sides. The method comprises receiving images from opposing sides of a vehicle, detecting a damage region in one image, and computing deep learning feature vectors for the damage region and a corresponding region in the opposite image. These fe...