Amazon. has been granted a patent for a computer-implemented method that enhances depth-guided structure from motion. The method involves obtaining image frames, performing correspondence searches for 2-D keypoints, and using a pretrained depth estimator to create dense depth maps and 3-D points, optimizing camera poses and refining point clouds. GlobalData’s report on Amazon gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on Amazon, Dynamic premium pricing was a key innovation area identified from patents. Amazon's grant share as of June 2024 was 82%. Grant share is based on the ratio of number of grants to total number of patents.
Depth-guided structure from motion for 3-d point generation
The granted patent US12046002B1 outlines a computer-implemented method and system for processing a series of image frames from a digital content item to generate a 3-D point cloud and optimize camera poses. The method begins by obtaining multiple image frames corresponding to a scene and performing a correspondence search to identify matched 2-D keypoints across these frames. A pretrained depth estimator is then utilized to create dense depth maps, which are used to derive depth priors for the identified keypoints. The process includes determining 3-D points based on the 2-D keypoints, depth priors, and intrinsic matrices, followed by the selection of image frame pairs to establish initial camera poses. The method further refines these poses and adds additional 3-D points to the point cloud through triangulation, optimizing the results based on reprojection and depth consistency errors.
Additionally, the patent specifies that the system comprises processors and memory that execute instructions to carry out the aforementioned steps. It highlights the use of bilinear interpolation for extracting depth priors from dense depth maps and emphasizes the importance of minimizing errors during the optimization process. The claims also mention that a majority of the image frames should have a parallax of 1.0 or less, which is significant for ensuring accurate depth estimation. The patent encompasses various aspects, including the initialization of image frame pairs based on 2-D keypoint correspondences and the incremental updating of camera poses, thereby providing a comprehensive framework for enhancing 3-D reconstruction from 2-D image data.
To know more about GlobalData’s detailed insights on Amazon, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
