公开(公告)号：US11481683B1

公开(公告)日：2022-10-25

申请号：US16888589

申请日：2020-05-29

Applicant: Amazon Technologies, Inc.

Inventor： Kunwar Yashraj Singh ,  Joaquin Zepeda Salvatierra ,  Erhan Bas ,  Vijay Mahadevan ,  Jonathan Wu ,  Rahul Bhotika

IPC: G06F15/18 ,  G06N20/00 ,  G06T3/00 ,  G06N5/04 ,  G06F17/16

Abstract: Techniques for creating machine learning models for direct homography regression for image rectification are described. In certain embodiments, a training service trains an algorithm on a source view of a training image and a homography matrix of the training image into a machine learning model that generates a normalized homography matrix for an input of the source view. The normalized homography matrix may then be utilized to generate a target view of an image input into the machine learning model. The target view of the image may be used in a document processing pipeline for document images captured using cameras.