Reconstruction method of three-dimensional reconstruction model based on two-dimensional Gaussian splashing

By introducing a prunable anchor point parameterized framework and multi-scale feature fusion mechanism, the traditional two-dimensional Gaussian splashing in multi-view geometric consistency and storage overhead are solved, and lightweight and high-precision three-dimensional reconstruction is achieved, especially in textureless areas, which significantly improves the reconstruction effect.

CN120374867APending Publication Date: 2025-07-25GUANGDONG BOHUA UHD INNOVATION CENT CO LTD
0 Cites 26 Cited by

Patent Information

Application Number
CN202510521249.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-24
Publication Date
2025-07-25

AI Technical Summary

Technical Problem

The traditional two-dimensional Gaussian splashing method has shortcomings in multi-view geometric consistency, storage overhead and weak texture area reconstruction, and the adaptive mechanism is insufficient, making it difficult to meet the high-quality three-dimensional reconstruction needs of lightweight equipment.

Method used

A parameterized framework for prunable anchor points and a multi-scale feature fusion mechanism are introduced, and sparse three-dimensional point clouds are generated through multi-view stereoscopic vision algorithms, visible anchor points are dynamically filtered, and differentiable rendering pipelines and loss function optimization are used to build a compact scene representation model.

Benefits of technology

It improves reconstruction accuracy, reduces storage overhead, solves the problem of Gaussian distribution ambiguity in textureless areas, and provides a high-quality three-dimensional reconstruction solution for miniaturized devices.

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

Abstract

The invention provides a reconstruction method of a three-dimensional reconstruction model based on two-dimensional Gaussian splashing, which comprises the following steps: S1, carrying out sparse reconstruction on an input image sequence through a multi-view stereoscopic vision algorithm to generate an initial sparse three-dimensional point cloud and a corresponding camera pose parameter; s2, inputting an improved two-dimensional Gaussian radiation field by using the sparse three-dimensional point cloud and the camera pose as information; s3, dynamically screening a visible anchor point subset based on the current view angle parameter, and generating a rendered image through a differentiable rendering pipeline; s4, calculating a loss function of the rendering image of the training track and the input image to optimize a reconstruction scene; and S5, starting a special visualization tool, and inputting a rendering result. According to the method, by introducing a trimmable anchor point parameterization framework and a multi-scale feature fusion mechanism, light-weight and high-precision three-dimensional scene modeling is achieved, and the problems that traditional 2D Gaussian sputtering is insufficient in multi-view geometric consistency, storage overhead and weak texture region reconstruction and an existing 2D Gaussian splashing method is insufficient in self-adaptive mechanism are solved.
Need to check novelty before this filing date? Find Prior Art