Meshtryoshka enables differentiable rendering of large-scale scenes with standard mesh rasterizers by using nested mesh shells from an SDF with indirect gradient flow.
hub Canonical reference
NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction
Canonical reference. 100% of citing Pith papers cite this work as background.
38
Pith papers citing it
Background
100% of classified citations
abstract
We present a novel neural surface reconstruction method, called NeuS, for reconstructing objects and scenes with high fidelity from 2D image inputs. Existing neural surface reconstruction approaches, such as DVR and IDR, require foreground mask as supervision, easily get trapped in local minima, and therefore struggle with the reconstruction of objects with severe self-occlusion or thin structures. Meanwhile, recent neural methods for novel view synthesis, such as NeRF and its variants, use volume rendering to produce a neural scene representation with robustness of optimization, even for highly complex objects. However, extracting high-quality surfaces from this learned implicit representation is difficult because there are not sufficient surface constraints in the representation. In NeuS, we propose to represent a surface as the zero-level set of a signed distance function (SDF) and develop a new volume rendering method to train a neural SDF representation. We observe that the conventional volume rendering method causes inherent geometric errors (i.e. bias) for surface reconstruction, and therefore propose a new formulation that is free of bias in the first order of approximation, thus leading to more accurate surface reconstruction even without the mask supervision. Experiments on the DTU dataset and the BlendedMVS dataset show that NeuS outperforms the state-of-the-arts in high-quality surface reconstruction, especially for objects and scenes with complex structures and self-occlusion.
hub tools
citation-role summary
background 5
citation-polarity summary
roles
background 5polarities
background 5representative citing papers
GenRecon lifts object-level generative priors to scene-scale reconstruction by chunking scenes and using projection-based conditioning on multi-view features, claiming 16% better results than prior methods.
PAGaS refines multi-view stereo depths by optimizing 1DoF Gaussians whose positions and sizes are fixed by back-projected pixel volumes, producing detailed depth maps that outperform reference baselines on 3D reconstruction benchmarks.
SpUDD defines superpower contours from power diagrams of unsigned distance samples, proves convergence to the true surface, and uses them to generate approximating polygonal meshes that outperform prior strategies.
THOM is a training-free two-stage framework that generates physically plausible hand-object 3D meshes directly from text by combining text-guided Gaussians with contact-aware physics optimization and VLM refinement.
SVGS improves Gaussian Splatting novel-view synthesis by replacing single-color primitives with spatially varying color and opacity functions implemented via bilinear interpolation, movable kernels, or tiny neural networks on 2D Gaussian surfels.
HSDF-Lane uses a height-aligned signed distance field with differentiable rendering and lane-aware semantic positional encoding to achieve SOTA 3D lane detection and height estimation on OpenLane.
Four-stage pipeline reconstructs explicit facial hair strands from multi-view images via 3D Gaussian optimization and curve tracing, claimed as first such method.
SAGE self-learns Gaussian expression deformations via joint surfel-SDF optimization and self-supervised consistency, enabling comparable avatar quality from single frames, monocular rotations, or one-shot inputs.
MeshFlow uses a contrastive MeshVAE for compact mesh latents and a flow transformer for parallel generation, claiming 18x speedup over autoregressive methods with high accuracy on standard metrics.
GARDEN uses gravity alignment and conditional 3D point classification to factorize RGB reconstructions into explicit rigid bodies plus decoupled background for direct physics simulation.
SAR2Mesh performs coarse-to-fine 3D cardiac mesh reconstruction from SAR radar by template deformation with geometry-aware projection and physics-informed loss, outperforming baselines on the introduced Cardiac Mesh-SAR dataset.
ArtMesh presents a mesh-native pipeline for articulated reconstruction that uses restricted Delaunay remeshing and bidirectional motion consistency to outperform 3D Gaussian Splatting methods on joint estimation and part geometry.
TOPOS creates high-fidelity 3D heads with fixed industry topology from single images via a specialized VAE with Perceiver Resampler and a rectified flow transformer.
RetrieveVGGT enables constant-memory long-context streaming 3D reconstruction by retrieving relevant frames via query-key similarities in VGGT's first attention layer, outperforming StreamVGGT and others.
A framework that structurally enforces divergence-free velocity and long-range transport coherence in 3D fluid reconstruction from 2D videos via divergence-free kernels advecting Lagrangian Gaussian splats.
Sat3R adapts Depth Anything V2 via RPC-aware metric depth fine-tuning to deliver satellite DSM reconstruction with 38% lower MAE than zero-shot baselines and over 300x speedup versus optimization methods.
A new GPU clipping algorithm with directional culling and hierarchical traversal constructs scalable 3D Voronoi and power diagrams for arbitrary point distributions.
A single-image head reconstruction method uses coarse-to-fine optimization with normal consistency, landmarks, and geometry-aware constraints on curvature and conformality to produce meshes with industry-grade topology and preserved facial identity.
A greedy algorithm interpolates consistent signed distance functions from discrete samples by treating SDF geometric properties as hard constraints.
A feed-forward model regresses accurate Gaussian surfel geometry from sparse views using Nyquist-guided cross-view feature aggregation, achieving 100x speedup over optimization-based 3DGS surface methods on DTU benchmarks.
Neural Harmonic Textures add periodic feature interpolation and deferred neural decoding to primitive representations, achieving state-of-the-art real-time novel-view synthesis and bridging primitive and neural-field methods.
A pipeline that reconstructs articulated objects from sparse unposed images by aligning independent per-pose reconstructions via learned deformation fields and progressive static/moving part disentanglement.
EpiS improves generalizable neural surface reconstruction from sparse views by guiding epipolar feature aggregation with cost volumes, using an epipolar transformer, and applying pretrained monocular depth constraints, outperforming prior methods on DTU and BlendedMVS.
citing papers explorer
-
PREF: Phasorial Embedding Fields for Compact Neural Representations
PREF introduces a phasor volume and tailored Fourier mapping to let shallow MLPs capture high-frequency signals compactly in 2D images, 3D SDFs, and 5D NeRFs.