REVIEW 26 cited by
Implicit Geometric Regularization for Learning Shapes
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Representing shapes as level sets of neural networks has been recently proved to be useful for different shape analysis and reconstruction tasks. So far, such representations were computed using either: (i) pre-computed implicit shape representations; or (ii) loss functions explicitly defined over the neural level sets. In this paper we offer a new paradigm for computing high fidelity implicit neural representations directly from raw data (i.e., point clouds, with or without normal information). We observe that a rather simple loss function, encouraging the neural network to vanish on the input point cloud and to have a unit norm gradient, possesses an implicit geometric regularization property that favors smooth and natural zero level set surfaces, avoiding bad zero-loss solutions. We provide a theoretical analysis of this property for the linear case, and show that, in practice, our method leads to state of the art implicit neural representations with higher level-of-details and fidelity compared to previous methods.
Forward citations
Cited by 26 Pith papers
-
Points as Tori: Fast Pointwise Signed Distance for Point Clouds
Blending closed-form torus SDFs, with per-point coefficients predicted by a shared neural network, yields pointwise signed distance to point clouds without explicit reconstruction.
-
Geometry Field Splatting with Gaussian Surfels
Gaussian surfels can parameterize a stochastic geometry field with a closed-form, near-exact differentiable splatting renderer, improving 3D surface reconstruction on DTU and BlendedMVS.
-
Masked Topology Modeling for Self-Supervised Learning on Parametric CAD
Masked Topology Modeling pretrains B-rep encoders by hiding face-adjacency edges and predicting their kernel-computed convexity and curve type, improving label efficiency on CAD benchmarks.
-
Hydrodynamic Effects in Cryogenic Buffer Gas Cells: Design Insights from Hybrid Simulations
In a spherical cryogenic buffer gas cell, vortices form at certain helium flows and injection angles and can enhance molecule extraction, which should be visible in beam velocity or time-of-flight measurements.
-
Mechanics Simulation with Implicit Neural Representations of Complex Geometries
A framework that uses neural implicit geometry representations to feed shifted-boundary finite element simulations, removing explicit surface meshing for linear elasticity on complex shapes.
-
ViscoReg: Neural Signed Distance Functions via Viscosity Solutions
A viscosity-regularized Eikonal loss with annealed epsilon improves Neural SDF reconstruction and yields the first generalization bound for SDF learning.
-
NOVA3D: Normal Aligned Video Diffusion Model for Single Image to 3D Generation
A video diffusion model fine-tuned to output both color and normal maps, aligned by a geometry-temporal attention block, reconstructs textured 3D meshes from a single image.
-
Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space
Common3D learns deformable 3D morphable models for common objects from object-centric videos using neural features, and solves pose, segmentation, and semantic correspondence in a self-supervised, zero-shot manner.
-
Creating Your Editable 3D Photorealistic Avatar with Tetrahedron-constrained Gaussian Splatting
TetGS is a hybrid representation that embeds Gaussian kernels inside tetrahedral grids, enabling locally controlled geometric and appearance edits of 3D avatars reconstructed from monocular video.
-
Joint Optimization of Neural Radiance Fields and Continuous Camera Motion from a Monocular Video
A NeRF-based system estimates camera motion as continuous velocities and recovers accurate camera poses and scene geometry from monocular video without depth priors.
-
Few-Shot Multi-Human Neural Rendering Using Geometry Constraints
A neural implicit method uses SMPL body meshes as geometric priors to reconstruct and re-render multi-human scenes from 5 to 20 input views.
-
Robustifying Fourier Features Embeddings for Implicit Neural Representations
A bias-free MLP filter applied multiplicatively to Fourier features, with a line-search learning-rate controller, reduces noise and improves implicit neural representation fitting across images, shapes, and NeRF.
-
Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views
Inter-image feature matching, not monocular depth, supplies the depth prior that enables accurate neural implicit surface reconstruction from sparse indoor views.
-
DGNS: Deformable Gaussian Splatting and Dynamic Neural Surface for Monocular Dynamic 3D Reconstruction
A hybrid of deformable Gaussian splatting and dynamic neural SDF achieves state-of-the-art 3D mesh accuracy from monocular video while keeping view synthesis competitive.
-
Neural 4D Evolution under Large Topological Changes from 2D Images
N4DE learns 4D deformations with large topology changes from 2D images by evolving a time-conditioned neural SDF with hash grids and implicit Gaussian splatting.
-
Robust SG-NeRF: Robust Scene Graph Aided Neural Surface Reconstruction
A pose-confidence method using a detached view-direction-free color network, Monte Carlo re-localization, and dynamic scene graph updates improves neural surface reconstruction under outlier camera poses.
-
Spacecraft Safe Robust Control Using Implicit Neural Representation for Geometrically Complex Targets in Proximity Operations
A two-layer robust controller uses a learned signed distance function with conservative error margins and circulation inequalities to keep a chaser spacecraft collision-free near complex-geometry targets.
-
HiNeuS: High-fidelity Neural Surface Mitigating Low-texture and Reflective Ambiguity
HiNeuS builds accurate 3D surfaces from photos by combining SDF-based visibility checks, local planar regularization, and rendering-error-weighted Eikonal constraints, reporting SOTA on several benchmarks.
-
3D Surface Reconstruction with Enhanced High-Frequency Details
FreNeuS adds gradient-guided ray sampling and high-frequency-weighted color loss to NeuS, reducing mean Chamfer distance from 0.84 to 0.73 on DTU and from 1.97 to 1.11 on six NeRF-synthetic scenes.
-
Spatially-Adaptive Hash Encodings For Neural Surface Reconstruction
A learned spatial mask modulates multi-resolution hash grid features per scene location, improving neural surface reconstruction accuracy on DTU and Tanks and Temples.
-
HotSpot: Signed Distance Function Optimization with an Asymptotically Sufficient Condition
A heat-equation-based loss for neural signed distance functions gives an asymptotically sufficient condition for convergence to the true distance, with better surface and distance accuracy on shape benchmarks.
-
Neural shape reconstruction from multiple views with static pattern projection
A neural signed-distance field is jointly optimized with camera and projector poses to fuse structured-light images captured from freely moving devices.
-
SparSplat: Fast Multi-View Reconstruction with Generalizable 2D Gaussian Splatting
A feed-forward model regressing 2D Gaussian splat parameters from three views reports the best Chamfer distance on DTU sparse reconstruction in its comparison table, and competitive novel view synthesis, at roughly 80...
-
NumGrad-Pull: Numerical Gradient Guided Tri-plane Representation for Surface Reconstruction from Point Clouds
NumGrad-Pull uses tri-plane feature grids, numerical gradients, progressive resolution increases, and complementary sampling to improve signed-distance-function surface reconstruction from point clouds.
-
$\text{S}^{3}$Mamba: Arbitrary-Scale Super-Resolution via Scaleable State Space Model
S3Mamba applies scale-modulated state space models to arbitrary-scale super-resolution, reporting marginal PSNR gains over prior INR-based methods.
-
STITCH: Surface reconstrucTion using Implicit neural representations with Topology Constraints and persistent Homology
STITCH augments Neural-Pull with a topological loss derived from persistent homology to encourage a single connected component in reconstructed surfaces.
Discussion (0). Continue with ORCID to comment.