Sparse view angle three-dimensional reconstruction method and system based on voxel grid constraint

By employing a multi-view stereo geometric model and a voxel mesh constraint optimization strategy, the problem of reconstruction quality degradation in 3D Gaussian splashes under sparse perspectives was solved, achieving high-precision 3D reconstruction under sparse perspectives, which is suitable for robot perception and virtual reality.

CN121527352APending Publication Date: 2026-02-13BEIJING INST OF TECH
1 Cites 3 Cited by

Patent Information

Application Number
CN202511569327.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-10-30
Publication Date
2026-02-13

AI Technical Summary

Technical Problem

Existing 3D Gaussian splashing methods suffer from quality degradation in sparse viewpoints, producing floating objects or artifacts, and rely on high-quality prior information, resulting in insufficient robustness.

Method used

Camera parameters and dense 3D point clouds are obtained through a multi-view 3D geometric model. Voxel meshes are constructed and point cloud geometric features and confidence scores are fused. Gradient decay and mesh control strategies for voxel mesh constraints are designed to optimize the 3D Gaussian splash radiation field.

Benefits of technology

It significantly improves reconstruction accuracy under sparse perspectives, reduces artifact generation, enhances the robustness of the method, and is suitable for applications such as robot perception and virtual reality.

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

Abstract

The invention discloses a sparse visual angle three-dimensional reconstruction method and system based on voxel grid constraint, and belongs to the technical field of visual three-dimensional reconstruction. Constructing a three-dimensional voxel grid of a self-adaptive scene scale based on point cloud distribution, and dividing the point cloud to the corresponding voxel grid; fusing the geometric features of the fast point feature histogram and the confidence score in the grid, generating a geometric confidence comprehensive measure, and screening key points to initialize Gaussian primitives; designing a voxel grid constrained gradient clipping strategy, limiting Gaussian primitive error diffusion through a distance attenuation coefficient, and adaptively optimizing grid distribution in combination with a dynamic grid deletion and addition mechanism; and finally, carrying out iterative training by using a 3D Gaussian splash radiation field loss function to realize high-fidelity static scene reconstruction under a sparse view angle. According to the method, scene geometric priori is introduced, an optimization strategy based on voxel grid constraint is designed to effectively control excessive diffusion or drift of Gaussian primitives, generation of artifacts is reduced, and meanwhile, the situation that robustness is reduced due to the influence of priori quality is avoided.
Need to check novelty before this filing date? Find Prior Art