Depth estimation for SLAM systems using monocular cameras

CN119365892APending Publication Date: 2025-01-24INNOPEAK TECHNOLOGY INC
0 Cites 0 Cited by

Patent Information

Application Number
CN202280095249.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-04-27
Publication Date
2025-01-24

AI Technical Summary

Technical Problem

When performing depth image estimation from a monocular mobile camera in augmented reality applications, there are problems such as high computational burden, high noise and low resolution, and the SLAM-based depth estimation module needs to be trained on a large number of test data sets. , resulting in inefficiency.

Method used

By obtaining the current image and reference image, the pixel -based search table is used to determine the depth of each pixel to avoid polar correction and deep learning technology, and improve the speed and accuracy of deep estimates.

Benefits of technology

The depth information is determined quickly and accurately from the image data, which is suitable for expanding the actual application, reducing the calculation burden and noise, and improving the efficiency and resolution of deep estimates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119365892A_ABST
    Figure CN119365892A_ABST
Patent Text Reader

Abstract

An electronic device acquires a current image and a reference image captured by a camera device. Each current pixel in at least one subset of the current image has a current pixel position in the current image and corresponds to a corresponding reference pixel in the reference image. For each current pixel in at least the subset of a current image, the electronic device obtains a pixel conversion parameter of the current pixel, and determines a parallax between a current distance between the current pixel of the current image and a current pole and a reference distance between a corresponding reference pixel of the reference image and a reference pole. And according to a pixel conversion parameter, the electronic equipment determines a depth value of the current pixel from the parallax between the current distance and the reference distance, and creates a depth map corresponding to the current image.
Need to check novelty before this filing date? Find Prior Art