paper-with-me

홈 › Papers

VoxNeuS: Enhancing Voxel-Based Neural Surface Reconstruction via Gradient Interpolation

2024-06-11 · Sidun Liu, Peng Qiao, Zongxin Ye, Wenyu Li, Yong Dou

Neural Surface Reconstruction learns a Signed Distance Field~(SDF) to reconstruct the 3D model from multi-view images. Previous works adopt voxel-based explicit representation to improve efficiency. However, they ignored the gradient instability of interpolation in the voxel grid, leading to degradation on convergence and smoothness. Besides, previous works entangled the optimization of geometry and radiance, which leads to the deformation of geometry to explain radiance, causing artifacts when reconstructing textured planes. In this work, we reveal that the instability of gradient comes from its discontinuity during trilinear interpolation, and propose to use the interpolated gradient instead of the original analytical gradient to eliminate the discontinuity. Based on gradient interpolation, we propose VoxNeuS, a lightweight surface reconstruction method for computational and memory efficient neural surface reconstruction. Thanks to the explicit representation, the gradient of regularization terms, i.e. Eikonal and curvature loss, are directly solved, avoiding computation and memory-access overhead. Further, VoxNeuS adopts a geometry-radiance disentangled architecture to handle the geometry deformation from radiance optimization. The experimental results show that VoxNeuS achieves better reconstruction quality than previous works. The entire training process takes 15 minutes and less than 3 GB of memory on a single 2080ti GPU.

📄 PDF Abstract BibTeX arXiv:2406.07170

Code (0)

등록된 구현이 없습니다.

Tasks

GPUSurface Reconstruction

Similar Papers 제목 키워드 기반

Gradient-SDF: A Semi-Implicit Surface Representation for 3D Reconstruction

2021-11-26 · CVPR 2022 1 · Christiane Sommer, Lu Sang, David Schubert, Daniel Cremers

We present Gradient-SDF, a novel representation for 3D geometry that combines the advantages of implict and explicit representations. By storing at every voxel both the signed distance field as well as its gradient vecto…

3D geometry3D ReconstructionGPU

Inner-Outer Aware Reconstruction Model for Monocular 3D Scene Reconstruction

2023-09-21 · NeurIPS 2023 11

Monocular 3D scene reconstruction aims to reconstruct the 3D structure of scenes based on posed images. Recent volumetric-based methods directly predict the truncated signed distance function (TSDF) volume and have achie…

Sparse-Up: Learnable Sparse Upsampling for 3D Generation with High-Fidelity Textures

2025-09-28 · Lu Xiao, Jiale Zhang, Yang Liu, Taicheng Huang 외 arxiv

The creation of high-fidelity 3D assets is often hindered by a 'pixel-level pain point': the loss of high-frequency details. Existing methods often trade off one aspect for another: either sacrificing cross-view consiste…

3D Generation

GeoSVR: Taming Sparse Voxels for Geometrically Accurate Surface Reconstruction

2025-09-22 · Jiahe Li, Jiawei Zhang, Youmin Zhang, Xiao Bai 외 arxiv

Reconstructing accurate surfaces with radiance fields has achieved remarkable progress in recent years. However, prevailing approaches, primarily based on Gaussian Splatting, are increasingly constrained by representatio…

Voxurf: Voxel-based Efficient and Accurate Neural Surface Reconstruction

2022-08-26 · Tong Wu, Jiaqi Wang, Xingang Pan, Xudong Xu 외

Neural surface reconstruction aims to reconstruct accurate 3D surfaces based on multi-view images. Previous methods based on neural volume rendering mostly train a fully implicit model with MLPs, which typically require …

Surface Reconstruction