Supercharge Your Innovation With Domain-Expert AI Agents!

A method and system for rapidly generating building facade textures based on semantic understanding

A technology for building façade and semantic understanding, applied in the field of 3D reconstruction, can solve the problems of high repetition, long time consumption, large process, etc., and achieve the effect of low mapping efficiency and high similarity

Active Publication Date: 2022-05-27
北京飞渡科技股份有限公司
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method faces the following problems: (1) There are a large number of manual cutout processes and high repeatability; (2) It takes a very long time to manually repair walls, doors, and windows

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method and system for rapidly generating building facade textures based on semantic understanding
  • A method and system for rapidly generating building facade textures based on semantic understanding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] See Figure 1 , a rapid method for generating textures on building facades based on semantic understanding, including the following steps:

[0043] Step (1): Build a texture map library containing different first semantic component maps, in order to quickly build an elevation texture with high similarity and high realism, you need to prepare a texture map library of semantic categories in advance, such as wall texture map library, door texture map library, window texture map library, etc. The texture map in the texture map library has the corresponding first semantic label and has a high degree of realism.

[0044] Step (2): Extract the original texture map of the 3D model façade that is automatically reconstructed.

[0045]Step (3): In order to automatically identify and divide the semantics of the original texture map, such as doors, windows, walls, etc., the semantic recognition network needs to be used, therefore, the semantic recognition network training set needs to be...

Embodiment 2

[0054] See Figure 2 , a rapid generation system for building façade textures based on semantic understanding, including: acquisition module, for scanning according to different physical components to obtain the original image of the physical components; processing module, based on the original image to configure the corresponding parameter library to generate component maps; writing modules, based on the component maps and writing the corresponding first semantics in the first attribute value of the component maps; texture map library, for storing component maps with different first semantics in the corresponding storage unit to form a texture map library Identification module, which is used to extract the original texture map of the 3D model façade and recognize the second semantics of the original texture map; the label module, which is used to write a label representation in the second attribute value of the original texture map based on the second semantic map of the original ...

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

No PUM Login to View More

Abstract

The invention discloses a method and system for quickly generating building facade textures based on semantic understanding, and belongs to the technical field of three-dimensional reconstruction. The method includes: S1: building a texture map library containing component maps with different first semantics; S2: extracting the original texture map of the facade of the three-dimensional model; S3: identifying the second semantics of the original texture map, and obtaining the corresponding first texture map Two semantic label maps; S4: Clipping the original texture map through the second semantic label map to obtain specific semantic clipping images corresponding to different parts of the 3D model; S5: Retrieve the texture map corresponding to the clipping clipping images in the texture map library. In the present invention, based on semantic understanding and automatic retrieval methods, on the basis of semantic analysis of doors, windows, walls, etc., the most similar textures are retrieved from the texture library and replaced to construct textures with high similarity and high authenticity. The texture of the façade is more realistic, thus solving the problem of low efficiency and long time for manually repairing the texture of the 3D reconstruction model.

Description

Technical field [0001] The present invention relates to the field of three-dimensional reconstruction technology, particularly to a method and system for rapid generation of building façade textures based on semantic understanding. Background [0002] In the traditional building map making industry, the texture on the three-dimensional model of the building automatically obtained through multi-mesh reconstruction often has the problem of distortion, distortion and blur, and the prior art will first manually deduct the doors, windows and other components, and then manually repair the walls first, and then manually repair the doors, windows and other components. This method faces the following problems: (1) the process of manual keying is large and repetitive; (2) the manual repair of walls, doors, and windows takes a long time. Contents of the Invention [0003] In order to solve the problem of prior art, the present invention provides a method and system for rapid generation of ...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T17/00G06T15/04G06N20/00
CPCG06T17/00G06T15/04G06N20/00
Inventor 何文武宋彬朱旭平
Owner 北京飞渡科技股份有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com