A method for precise positioning of a non-positioned photo based on a real scene three-dimensional model

By generating a virtual image database based on a real-world 3D model and training a feature extractor, the problem of image localization without prior information is solved, achieving high-precision and robust image spatial localization, which is suitable for emergency response and public safety scenarios.

CN122176056APending Publication Date: 2026-06-09PERCEPTION WORLD (BEIJING) INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
PERCEPTION WORLD (BEIJING) INFORMATION TECH CO LTD
Filing Date
2026-03-18
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

In emergency response and public safety scenarios, the lack of reliable image metadata and attitude parameters makes image spatial positioning difficult. Traditional methods have insufficient positioning accuracy or poor robustness without prior information.

Method used

A virtual image database is generated based on a real-scene 3D model. A feature extractor is trained using an OTFAM model, a feature library is constructed, and feature matching and pose calculation are performed to achieve high-precision and robust image localization.

Benefits of technology

High-precision spatial positioning of images was achieved without position and attitude information, improving the recall and robustness of positioning and enabling it to meet the image positioning needs under complex shooting conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122176056A_ABST
    Figure CN122176056A_ABST
Patent Text Reader

Abstract

This invention relates to the fields of computer vision and measurement technology, and proposes a method for precise localization of images without location information based on a real-world 3D model. The method includes four steps: Step 1: virtual projection image generation based on the real-world 3D model; Step 2: feature extractor construction and training; Step 3: feature library construction and feature matching; and Step 4: image pose calculation after retrieval. The virtual projection image generation method uses a real-world 3D model as a foundation, sets up a virtual camera according to preset observation distance, viewing angle range, and attitude parameters, performs multi-view virtual projection, and generates model image data associated with spatial location and height information. This invention requires no known shooting pose or spatial prior conditions, relying solely on the real-world 3D model as a reference data source to achieve high-precision spatial localization of images, filling the technical gap of being unable to locate images without location information, and greatly expanding the range of sources of locatable images.
Need to check novelty before this filing date? Find Prior Art