Method and apparatus for constructing XR scene layered spatial maps based on scene element features

By dividing scene elements into dynamic features, positional features, and interactive features, a layered spatial map is constructed, which solves the adaptability problem of SLAM technology in dynamic scenes and improves the real-time perception and virtual reality fusion effect of XR devices.

CN121120966BActive Publication Date: 2026-06-30SHANGHAI CONDENSATION YUANJIE INFORMATION TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANGHAI CONDENSATION YUANJIE INFORMATION TECH CO LTD
Filing Date
2025-08-20
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing SLAM technology is not adaptable enough to dynamic scenes and cannot effectively distinguish the characteristics of scene elements. This makes it difficult to balance the accuracy of map building and the efficiency of updating, affecting the real-time perception of XR devices and the accurate integration of virtual content with real-world scenes.

Method used

By acquiring the dynamic, positional, and interactive features of elements in the actual scene, they are divided into three categories: static, quasi-static, and dynamic, and stored in different layers to construct a hierarchical spatial map, which is then displayed on the XR device. Storage structures such as octree, KD-Tree, and voxel grid are used for management.

Benefits of technology

It achieves improved update efficiency while ensuring map accuracy, supports real-time navigation for XR devices and accurate integration of virtual content with real-world scenes, and reduces waste of computing resources and storage costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121120966B_ABST
    Figure CN121120966B_ABST
Patent Text Reader

Abstract

This invention relates to the field of 3D map technology and discloses a method and apparatus for creating a layered spatial map of XR scenes based on scene element features. This solution addresses the problem of balancing accuracy and update efficiency when using XR devices for map management and scene perception. The method includes: using data acquisition sensors to acquire actual scene elements and associated virtual scene elements; acquiring the dynamic features of the actual scene elements, their positional features in the XR scene, and the interaction features of the virtual scene elements; classifying the actual scene elements into various dynamic categories based on the dynamic, positional, and interaction features; and storing the actual scene elements of different dynamic categories in storage structures corresponding to different storage levels, constructing and displaying a layered spatial map of the XR scene corresponding to the target scene on the XR device. The storage levels include a static layer, a quasi-static layer, and a dynamic layer.
Need to check novelty before this filing date? Find Prior Art