A vectorization reverse construction method of inventory building indoor structure scene

By combining the synchronous processing of laser point clouds and panoramic image sequences, and utilizing image semantic segmentation and 3D convolutional networks, the problem of distinguishing between building structures and non-building structures in existing building interior structural scenes has been solved. This has enabled highly accurate vectorized reverse construction, and the generated 2D structural drawings meet the needs of engineering applications.

CN122415797APending Publication Date: 2026-07-17WUHAN UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
WUHAN UNIV
Filing Date
2026-03-26
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately distinguish between building structures and non-building structures in existing building interior structural scenarios, resulting in insufficient geometric continuity and structural semantic confusion in the reconstruction results. The generated two-dimensional structural drawings often have incomplete wall closures, missing doors and windows, or distorted structural boundaries, failing to meet actual needs.

Method used

By simultaneously acquiring laser point cloud data and panoramic image sequences, a pixel-by-pixel semantic discrimination is performed using an image semantic segmentation network to generate structural semantic point clouds. An initial semantic voxel field is constructed, non-building structure categories are eliminated, and virtual site cloud is generated. Combining voxelized block completion and 3D convolutional completion networks, a 3D structural mesh model is generated and vectorized.

Benefits of technology

It significantly improves the accuracy of vectorized reverse engineering of interior structural scenes of existing buildings. The generated two-dimensional structural vector drawings are superior to traditional methods in terms of topological closure, structural accuracy and engineering usability, and can directly support design comparison and engineering calculation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415797A_ABST
    Figure CN122415797A_ABST
Patent Text Reader

Abstract

The application provides a vectorization reverse construction method of an existing building indoor structure scene, and relates to the field of data processing. In the method, laser point cloud data and panoramic image sequences are synchronously acquired, and building structure semantics are extracted by image semantic segmentation. The image semantics are robustly migrated to the point cloud to form structure semantic point clouds. On the basis of eliminating non-building structure elements, a structure point cloud skeleton is constructed, and virtual placeholder point clouds are generated for open structures such as doorways and window openings. Then, continuous structure geometry is recovered by combining voxelization blocking completion and a three-dimensional convolution completion network. Finally, a three-dimensional structure grid model binding structure semantics is generated, and a two-dimensional structure vector drawing meeting engineering drawing rules is obtained by projection vectorization. The technical solution provided by the application facilitates improving the accuracy of the vectorization reverse construction of the existing building indoor structure scene.
Need to check novelty before this filing date? Find Prior Art