Client-based point cloud map update optimization method, system, medium and equipment
An update method and client-side technology, applied in the computer field, can solve problems such as time-consuming, high cost, multiple redundant calculations, etc., to save time, ensure consistency, and reduce map updates.
Active Publication Date: 2021-10-01
上海影创信息科技有限公司
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
When the scene changes over time, it is necessary to re-collect the data and then build an offline map. It takes a long time to build a map each time. When the scene only changes locally, there is too much redundant calculation in such repeated map building. And it takes a lot of time to build a map each time, which is too expensive
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0041] The present invention provides a large-scale cloud map update and optimization method under multi-client, the cloud map based on SLAM algorithm contains map point cloud and key frame information, the large map is evenly divided into blocks, and the cloud map is used for Provide initial positioning and positioning update functions for the client (AR glasses). However, due to changes in the actual scene, the cloud map should also be updated accordingly.
[0042] The present invention includes two parts: the selection of new map points in the client and the optimized update of the new map in the cloud.
[0043] like figure 2 , the specific steps for selecting a new map point in the client include:
[0044] Step 1.1, extracting local map points and corresponding key frame feature point information in the client (comprising multiple clients), adopting the ORB feature extraction algorithm to perform feature extraction;
[0045] Step 1.2, match with the key frame of the cor...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a client-based point cloud map optimization update method, system, medium and equipment, including: Step 1: Select new map points in the client: extract local map points and corresponding key frame feature points in the client Information, match with the key frame of the corresponding area in the cloud map, select the map points that need to be updated according to the matching similarity, add the map points that need to be updated to the cloud map and mark the time; Step 2: Optimize and update the new map on the cloud : Confirm the sub-map block where the map point to be updated is located, and jointly optimize the new keyframe that can be seen by the new map point and the old map point to update the cloud map. The present invention updates the client map to the cloud map irregularly, which ensures the consistency of the same scene between the client and the cloud.
Description
technical field [0001] The present invention relates to the field of computer technology, in particular to a client-based point cloud map update optimization method, system, medium and equipment. Background technique [0002] At this stage, the process of creating a map on the server side of a visual SLAM-based MR application is: first collect data and then create an offline map (including key frame information, map points, and common-view relationships). When the scene changes over time, it is necessary to re-collect the data and then build an offline map. It takes a long time to build a map each time. When the scene only changes locally, there is too much redundant calculation in such repeated map building. And it takes a lot of time to build a map each time, which is too expensive. [0003] Patent document CN108007453A (application number: CN201711307139.0) discloses a point cloud-based map update method, device and electronic equipment. Wherein, the method includes: ea...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06T7/73G01C21/00G01C21/20
CPCG06T7/73G01C21/005G01C21/20G06T2207/10016G06V20/46G06V10/751
Inventor 周乐
Owner 上海影创信息科技有限公司