Method and system for carrying out spatial three-dimensional modeling by using mobile equipment

By using mobile device photography and semantic segmentation technology, three-dimensional point cloud data containing human body information is generated, which solves the problem of high manual labeling costs in existing technologies and realizes efficient three-dimensional modeling and human-computer interaction support.

CN120672968AActive Publication Date: 2025-09-19CHINA NAT POSTAL & TELECOMM APPLIANCES CORP +2
7 Cites -1 Cited by

Patent Information

Application Number
CN202511157452.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-19
Publication Date
2025-09-19
Estimated Expiration
2045-08-19

AI Technical Summary

Technical Problem

Existing 3D modeling methods are unable to automatically generate point cloud data with semantic label information and containing human body information, which makes it difficult to identify human body categories in human-computer interaction application scenarios and the cost of manual labeling is high.

Method used

Use mobile devices for spatial three-dimensional modeling. By shooting scene depth images, extracting key frames and camera poses, constructing three-dimensional spatial point cloud data, and obtaining image semantic labels through semantic segmentation models, projecting and mapping them to point cloud data, point cloud data containing human body information is generated.

Benefits of technology

It realizes the automatic generation of point cloud data with semantic label information and human body information, reduces work costs, improves the information content and accuracy of point cloud data, and supports subsequent human-computer interaction applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120672968A_ABST
    Figure CN120672968A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of three-dimensional modeling, and provides a method and system for spatial three-dimensional modeling by using a mobile device, and the method comprises the steps: obtaining a to-be-modeled spatial scene which at least comprises a simulation human body having an interaction relationship with an environment object; shooting the to-be-modeled space scene based on the mobile device to obtain a scene depth image; based on the scene depth image, extracting a key frame and a camera pose of the to-be-modeled space scene; constructing three-dimensional space point cloud data based on the key frame and the camera pose; collecting a to-be-modeled space scene according to the camera pose to obtain a scene color image, and performing semantic segmentation on the scene color image to obtain an image semantic tag; and projecting and mapping the image semantic tag to the three-dimensional space point cloud data to obtain the three-dimensional space point cloud data with the image semantic tag. The point cloud data with semantic label information and human body information can be automatically generated, and the working cost is reduced.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Environment semantic mapping method based on deep convolutional neural network

    CN109636905A

  • Deep learning-based method for constructing three-dimensional semantic map of indoor environment

    CN110243370A

  • Robot obstacle avoidance method based on virtual scene training

    CN110977966A

  • Real-time semantic map construction method based on semantic inverse depth filtering

    CN111325843A

  • Indoor mobile robot three-dimensional semantic map construction method

    CN115035260A