Three-dimensional scene reconstruction method and system based on monocular depth estimation

By constructing a closed-loop collaborative system that integrates hybrid attention depth coding, adaptive edge perception, geometric consistency constraints, and Gaussian ellipsoid scene reconstruction modules, the problems of insufficient adaptive capability and limited depth estimation accuracy in existing technologies are solved, and high-quality 3D scene reconstruction is achieved.

CN121482285AActive Publication Date: 2026-02-06HARBIN INST OF TECH
5 Cites 3 Cited by

Patent Information

Application Number
CN202511939124.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-22
Publication Date
2026-02-06
Estimated Expiration
2045-12-22

AI Technical Summary

Technical Problem

Existing 3D scene reconstruction methods based on monocular depth estimation suffer from problems such as insufficient system adaptability due to lack of deep coupling between modules, limited global feature capture capability, limited depth estimation accuracy, large depth estimation error in edge regions, and a single 3D representation form.

Method used

A hybrid attention-based deep encoding module, an adaptive edge-aware deep decoding module, a geometric consistency constraint optimization module, and a Gaussian ellipsoid scene reconstruction module are employed to form a deeply coupled closed-loop collaborative system. Through multi-scale feature extraction, autoregressive depth prediction, image gradient guidance, geometric consistency feedback, and Gaussian ellipsoid representation, high-quality 3D scene reconstruction is achieved.

Benefits of technology

It significantly improves depth estimation accuracy, reduces edge region errors, enhances the system's adaptive optimization capabilities, and generates high-quality 3D scene representations to meet diverse application needs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a three-dimensional scene reconstruction method and system based on monocular depth estimation, and belongs to the technical field of computer vision and three-dimensional reconstruction, and the method comprises the steps: carrying out the multi-scale feature coding of a monocular RGB image through a mixed attention depth coding module, and obtaining the hierarchical depth feature representation; carrying out autoregression depth decoding through a self-adaptive edge perception depth decoding module to generate an initial depth map; a depth confidence map is calculated through a geometric consistency constraint optimization module and is fed back to a coding module for iterative optimization, and a refined depth map is output; and three-dimensional Gaussian ellipsoid scene representation is constructed through the Gaussian ellipsoid scene reconstruction module. According to the invention, high-precision depth estimation and high-quality three-dimensional reconstruction are realized by constructing a depth-coupled closed-loop cooperative system.
Need to check novelty before this filing date? Find Prior Art