Pith. sign in

REVIEW 4 major objections 4 minor 39 references

Neural Surface Priors for Editable Gaussian Splatting

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Neural surface priors let mesh edits rewrite Gaussian Splatting scenes

desk verdict Useful per-face edit propagation for 3DGS via an orthonormal-basis transform, but the claimed support for scaling and shearing is not backed by the math. read the letter →

arxiv 2411.18311 v2 pith:KYF3LRXF submitted 2024-11-27 cs.CV

Jakub Szymkowiak , Weronika Jakubowska , Dawid Malarz , Weronika Smolak-Dyżewska , Maciej Zięba , Przemyslaw Musialski , Wojtek Pa{l}ubicki , Przemys{l}aw Spurek This is my paper · ORCID
classification cs.CV
keywords 3DGaussianSplattingneuralsigneddistancefieldtrianglesoupproxymesh-guidededitingnovelviewsynthesissurfacereconstructionscene
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a scene reconstructed by 3D Gaussian Splatting can be edited intuitively by manipulating a mesh, provided the Gaussians are first trained under a neural surface prior. The authors' pipeline first recovers a neural signed distance field with PermutoSDF, extracts a mesh, and then trains flat Gaussians whose opacity is conditioned on distance to that surface. Each Gaussian is encoded as a triangle in a proxy soup; when the user edits the mesh with the same number of faces, each proxy triangle is re-expressed in the local frame of its nearest face, and the per-face basis change propagates the edit to the appearance. The result is a way to apply standard mesh tools such as translation, rotation, bevel, or knife cuts, and even physics simulations, to 3DGS scenes while keeping render quality close to plain 3DGS. If correct, it closes the gap between 3DGS and mesh-based modeling pipelines without requiring retraining after an edit.

What carries the argument

The central mechanism is the reversible triangle-soup encoding of flat Gaussians together with a per-face orthonormal-basis transfer map. Each Gaussian with center $m$, rotation $R$, and scaling $s$ is represented as an ordered triangle $V = [v_0, v_1, v_2]$ with $v_0 = m$, $v_1 = s_1 r_1$, $v_2 = s_2 r_2$; this encoding is invertible, so new triangles yield new Gaussian parameters. To propagate an edit, the method finds the nearest face $W$ of the extracted mesh for each proxy triangle, builds an orthonormal basis $U$ from that face, builds the analogous basis $U'$ from the corresponding edited face, and applies the transform $T = U'U^{-1}$ via $V' = T(V - w_0) + w'_0$. The opacity conditioning $\Phi_\beta(f_\theta(x))$ and a normal-alignment regularizer keep the Gaussians flat and surface-aligned, which is what makes the triangle encoding faithful.

What would settle it

Take a mesh region and apply a non-uniform scale (stretch along one axis only) while preserving face count; if the rendered appearance does not stretch correspondingly, the orthonormal transfer $T = U'U^{-1}$ has failed to carry the edit, disproving the claim of faithful mesh-guided propagation.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a neural SDF can serve as a geometric prior that makes 3D Gaussian Splatting amenable to mesh-guided appearance editing. The authors show that by conditioning Gaussian opacity on the signed distance to a PermutoSDF surface, fixing one scaling axis so each Gaussian is flat, and sampling initial Gaussian centers from the extracted mesh, each Gaussian becomes a reversible triangle in a triangular proxy. A mesh edit is then transferred by associating each proxy triangle with its nearest original mesh face, computing the orthonormal frame of that face before and after the edit, and applying the rotation $T = U'U^{-1}$ to the proxy triangle. The updated triangle recovers new Gaussian position, orientation, and scale, so the entire appearance follows the user's mesh manipulation. Experiments on NeRF Synthetic and BlendedMVS show PSNR close to plain 3DGS and demonstrate translation, rotation, knife, bevel, randomize, and wind-driven reshaping edits.

Load-bearing premise

The method assumes that the user's mesh edit leaves the number of faces unchanged and that every proxy triangle can be matched to a unique nearest mesh face; any edit that remeshes, splits, or merges faces breaks the correspondence, and the transfer map is a rotation, so non-uniform scaling or shear of faces is not faithfully represented.

Editorial extensions

If this is right

  • A user can edit a 3DGS scene with standard mesh tools (translate, rotate, knife, bevel, randomize) and the appearance updates without any retraining or per-edit optimization.
  • Physics simulations computed on the mesh, such as wind moving a plant's branches, can be transferred frame by frame to the rendered appearance.
  • Mesh resolution can be lowered for editing speed without hurting render quality, since metrics stay stable across mesh resolutions from 200 to 1000.
  • The method supports topology-aware workflows such as rigging and finite-element simulation because the underlying mesh retains connectivity, unlike pure triangle-soup approaches.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the transfer map is a rotation, a natural testable extension is to replace $U'U^{-1}$ with an affine or per-axis scaling map; this would likely improve fidelity for stretching and shearing edits, but it is not what the paper claims.
  • The fixed-topology requirement suggests that future work could build a hierarchical or deformation-aware correspondence between the proxy and the mesh, which would let remeshing edits be handled; the paper explicitly leaves this open.
  • The dependence on a separate PermutoSDF prior implies that the editing fidelity is bounded by the quality of the neural surface reconstruction, so scenes with floating artifacts or disconnected components will show editing artifacts even when the 3DGS appearance is clean.
  • The admitted shadow limitation (a moved object keeps its original shadow) indicates that the method edits geometry placement but not lighting, so integrating relighting or shadow recomputation is an obvious next step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes a mesh-guided editing pipeline for 3D Gaussian Splatting. First, PermutoSDF recovers a neural SDF and an extracted mesh; then flat Gaussian kernels are trained with opacity conditioned on SDF distance and with initial positions sampled from the mesh. For editing, the Gaussians are converted into a triangle soup proxy, the user edits the extracted mesh, and for each proxy triangle a per-face transformation is computed from orthonormal bases attached to the original and edited mesh faces. The transformation is applied to the proxy triangle, and updated Gaussian parameters are recovered. Experiments report novel-view synthesis PSNR against several baselines, a mesh-resolution study, an initialization ablation, and qualitative editing and physics demonstrations.

Significance. If the propagation claim were corrected, the pipeline would be a plausible practical contribution: it combines a neural surface prior with flat Gaussians and a triangle soup proxy, and the code is released. The two-stage surface-guided training is clearly described, and the initialization ablation is useful. I do not see a circularity problem: the neural SDF is trained self-supervised on the same images, and the edit propagation is a geometric transform rather than a fitted target. However, the current formulation transmits only rigid rotations and translations of the local frame, not scaling or shear, and the editing evaluation is qualitative, so the advertised 'wider range of modifications' is not established.

major comments (4)
  1. [III-C, Eqs. (5)-(6)] The propagation transform T = U'U^{-1} is a rigid rotation, not a general deformation. Both U and U' are orthonormal bases built from normalized edge vectors and face normals, so T preserves lengths and angles. Consequently, a mesh face that is uniformly scaled or sheared in-plane leaves the basis U' equal to U (up to rotation of the first edge direction), and the recovered Gaussian scales s1 and s2 from Eq. (4) remain unchanged. This directly contradicts Section V's statement that 'transformations such as scaling, shearing, translation and rotation are well supported' and the Abstract's claim of 'a wider range of modifications'. The sentence that T 'transforms W into W'' is also mathematically false when W' differs from W by scaling or shear. Please replace the basis-to-basis map with an affine map between corresponding triangles (or otherwise justify how scaling and shear are transmitted), and add a dedicated quantitative scaling/shear experiment.
  2. [IV-B, Fig. 7] The editing results are entirely qualitative. There are no metrics for edit fidelity, no comparison with GaMeS, SuGaR, Mani-GS, or GSDeformer, and no user study; Tables I-III measure novel-view synthesis and initialization only. Since the paper's central contribution is mesh-guided editing, the evaluation should include at least one quantitative protocol, such as comparing an edited render produced by this method against a mesh-bound baseline and reporting PSNR/LPIPS, or reporting a geometric deformation error on a controlled edit.
  3. [I vs. III-C] The Introduction states that edits can be applied 'whether directly applied or after optional remeshing,' but Section III-C assumes the edited mesh has the same number of faces as the extracted mesh and relies on a nearest-face association to maintain correspondence. Remeshing generally changes the face count and breaks this correspondence, so the 'optional remeshing' claim is unsupported. Section V correctly lists fixed topology as a limitation and should replace the Introduction's broader promise.
  4. [V and Fig. 2] The paper claims to 'fully leverage' mesh topology, but the propagation step uses only per-face nearest-neighbor association and never reads mesh adjacency or connectivity. The method therefore does not demonstrate topology-aware operations such as rigging, region-based deformation, or connectivity-preserving propagation. Please either soften this claim or demonstrate an operation that actually exploits mesh connectivity.
minor comments (4)
  1. [V] The Conclusions duplicate the first limitation sentence: 'First, the underlying mesh must provide a structurally sound approximation of the scene geometry. First, the underlying mesh must provide a sufficiently consistent approximation...' Please keep only one formulation.
  2. [V] The phrase 'lightning-dependent effects' should be 'lighting-dependent effects'.
  3. [III-C, Eq. (4)] The indexing of the rotation-matrix columns is inconsistent: Section III-B says the normal is the first column of the rotation matrix, while Eq. (4) orders the recovered columns as the normal, the first edge direction, and the second in-plane direction. The triangle vertex definitions in Section III-C should state explicitly which columns correspond to the in-plane scaling axes.
  4. [III-B, Eq. (2)] The opacity function Phi_beta is described as a bell-shaped function, but the text does not define what happens when f_theta(x) is negative; a short sentence clarifying the range of the argument would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central edit-propagation chain is a direct geometric construction from mesh correspondences, not a fitted prediction or self-citation-forced result.

full rationale

The load-bearing derivation is Section III-C. A proxy triangle V is encoded from each flat Gaussian (m, R, s), associated with the nearest face W of the extracted mesh, and updated by V' = T(V - w0) + w0' with T = U'U^{-1} (Eq. 6), where U and U' are orthonormal bases of corresponding mesh faces. New Gaussian parameters are recovered from V' via the reversible formula in Eq. 4. This is a geometric map from an edit to the appearance; no parameter is fitted to a quantity and then reported as a prediction of the same quantity. The neural SDF prior (PermutoSDF [26]) is trained on the same images as the appearance model, but it is used as a regularizer/initializer (opacity conditioning Eq. 2, normal regularizer Eq. 3, surface sampling) and the editing output is not claimed to be an independent prediction of that SDF. The reliance on GaMeS [30] for the flat-Gaussian/triangle-soup encoding is a component reuse with independent content and does not force the paper's conclusion by self-citation. The paper's own limitation statements (fixed topology, unchanged shadows) further show that the remaining claims are conditional rather than definitional. A genuine correctness concern exists but is outside circularity: because U and U' orthonormal bases use normalized edges and the face normal, T is a rigid rotation, so in-plane scaling/shear of a mesh face is not transmitted to the proxy, despite Section V claiming such transformations are 'well supported.' That is an overclaim relative to Eq. 6, not a reduction of the derivation to its own inputs.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced; the triangle soup proxy and the neural SDF are taken from prior work. The main assumptions are the quality of the PermutoSDF prior, the nearest-face association, and the fixed-topology correspondence.

free parameters (4)
  • opacity sharpness beta = learned, not reported
    Appears in Eq. 2 and controls the alignment width of the opacity with the SDF surface.
  • flat-Gaussian scale epsilon = 1e-8
    Fixed to a negligible value to force 2D Gaussians, following GaMeS.
  • initialization sample count = 100,000
    Number of points sampled from the extracted mesh to initialize Gaussian centers; Table III shows sensitivity to initialization.
  • normal regularizer weight = not reported
    The loss in Eq. 3 has no stated coefficient, leaving a free hyperparameter in the training objective.
assumptions (4)
  • domain assumption The mesh extracted from PermutoSDF approximates the scene geometry well enough to guide Gaussian placement and editing.
    The whole pipeline depends on the quality of this surface prior; the paper's own limitation section notes issues with disconnected components or floating artifacts.
  • domain assumption Each proxy triangle can be assigned a unique nearest mesh face.
    The edit propagation in Section III-C requires a nearest-face lookup; ambiguous or changing associations would break the correspondence.
  • domain assumption User edits preserve the number of mesh faces and a one-to-one correspondence.
    Stated as a limitation in Section V: remeshing, face-count-changing edits, or topology refinement are not supported.
  • ad hoc to paper The orthonormal basis transform T = U'U^{-1} captures the intended deformation.
    This construction is introduced in this paper; it is a rotation that maps old face directions to new face directions and discards length and shear information.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Neural Surface Priors for Editable Gaussian Splatting." pith.science (2026). https://pith.science/paper/KYF3LRXF

@misc{pith2026241118311,
  author       = {Pith},
  title        = {Pith review of: Neural Surface Priors for Editable Gaussian Splatting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KYF3LRXF}},
  note         = {Machine review of arXiv:2411.18311}
}
read the original abstract

In computer graphics and vision, recovering easily modifiable scene appearance from image data is crucial for applications such as content creation. We introduce a novel method that integrates 3D Gaussian Splatting with an implicit surface representation, enabling intuitive editing of recovered scenes through mesh manipulation. Starting with a set of input images and camera poses, our approach reconstructs the scene surface using a neural signed distance field. This neural surface acts as a geometric prior guiding the training of Gaussian Splatting components, ensuring their alignment with the scene geometry. To facilitate editing, we encode the visual and geometric information into a lightweight triangle soup proxy. Edits applied to the mesh extracted from the neural surface propagate seamlessly through this intermediate structure to update the recovered appearance. Unlike previous methods relying on the triangle soup proxy representation, our approach supports a wider range of modifications and fully leverages the mesh topology, enabling a more flexible and intuitive editing process. The complete source code for this project can be accessed at: https://github.com/WJakubowska/NeuralSurfacePriors.

Figures

Figures reproduced from arXiv: 2411.18311 by the authors.

Figure 1
Figure 1. We propose a novel method for editing 3D Gaussian Splatting (3DGS) scenes using an intermediate triangle soup proxy. Our approach leverages a [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Schematic overview of our pipeline. (1) Starting with a collection of input images and corresponding camera poses, the initial stage utilizes [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Visualization of propagating the mesh edit to a single proxy triangle. We start by connecting the triangle [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: A modification created with our method on a low-resolution mesh. The [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Comparison of rendering outputs for different Gaussian initialization [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Progression of a wind-driven simulation of a Ficus plant. Top row: changes in the object’s geometry. Bottom row: corresponding modifications in [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Example modifications produced with our method for scenes in the [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 15 canonical work pages

  1. [1]

    Mip-NeRF: A Multi- scale Representation for Anti-Aliasing Neural Radiance Fields

    Jonathan T. Barron et al. “Mip-NeRF: A Multi- scale Representation for Anti-Aliasing Neural Radiance Fields”. In: ICCV (2021)

    work page 2021

  2. [2]

    TensoRF: Tensorial Radiance Fields

    Anpei Chen et al. TensoRF: Tensorial Radiance Fields

    work page

  3. [3]

    NeuSG: Neural Implicit Surface Reconstruction with 3D Gaus- sian Splatting Guidance

    Hanlin Chen, Chen Li, and Gim Hee Lee. NeuSG: Neural Implicit Surface Reconstruction with 3D Gaus- sian Splatting Guidance . 2023. arXiv: 2312 . 00846 [cs.CV]. URL: https://arxiv.org/abs/2312.00846

    arXiv 2023

  4. [4]

    High-quality Surface Reconstruction using Gaussian Surfels

    Pinxuan Dai et al. High-quality Surface Reconstruc- tion using Gaussian Surfels . 2024. arXiv: 2404.17774 [cs.CV]. URL: https://arxiv.org/abs/2404.17774

    work page Pith review arXiv 2024

  5. [5]

    V olume rendering

    Robert A. Drebin, Loren Carpenter, and Pat Hanrahan. “V olume rendering”. In: Proceedings of the 15th An- nual Conference on Computer Graphics and Interactive Techniques. SIGGRAPH ’88. New York, NY , USA: As- sociation for Computing Machinery, 1988, pp. 65–74. ISBN : 0897912756. DOI: 10.1145/54852.378484. URL: https://doi.org/10.1145/54852.378484

    work page arXiv 1988

  6. [6]

    Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering

    Yutao Feng et al. Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering . 2024. arXiv: 2401.15318 [cs.GR]. URL: https://arxiv.org/ abs/2401.15318

    arXiv 2024

  7. [7]

    Mani-GS: Gaussian Splatting Ma- nipulation with Triangular Mesh

    Xiangjun Gao et al. Mani-GS: Gaussian Splatting Ma- nipulation with Triangular Mesh . 2024. arXiv: 2405 . 17811 [cs.GR]. URL: https : / / arxiv. org / abs / 2405 . 17811

    work page 2024

  8. [8]

    Multi-View Stereo Revisited

    M. Goesele, B. Curless, and S.M. Seitz. “Multi-View Stereo Revisited”. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recog- nition (CVPR’06) . V ol. 2. 2006, pp. 2402–2409. DOI: 10.1109/CVPR.2006.199

    work page doi:10.1109/cvpr.2006.199 2006

Show all 39 references
  1. [9]

    Gaussian Frost- ing: Editable Complex Radiance Fields with Real-Time Rendering

    Antoine Guédon and Vincent Lepetit. Gaussian Frost- ing: Editable Complex Radiance Fields with Real-Time Rendering. 2024. arXiv: 2403.14554 [cs.CV]. URL: https://arxiv.org/abs/2403.14554

    2024 arXiv

  2. [10]

    Sugar: Surface- aligned gaussian splatting for efficient 3d mesh recon- struction and high-quality mesh rendering

    Antoine Guédon and Vincent Lepetit. “Sugar: Surface- aligned gaussian splatting for efficient 3d mesh recon- struction and high-quality mesh rendering”. In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 2024, pp. 5354–5363

    2024

  3. [11]

    2D Gaussian Splatting for Geomet- rically Accurate Radiance Fields

    Binbin Huang et al. “2D Gaussian Splatting for Geomet- rically Accurate Radiance Fields”. In: Special Interest Group on Computer Graphics and Interactive Tech- niques Conference Conference Papers ’24. SIGGRAPH ’24. ACM, July 2024, pp. 1–11. DOI: 10.1145/3641519. 3657428. URL: ...

    2024 doi

  4. [12]

    SC-GS: Sparse-Controlled Gaus- sian Splatting for Editable Dynamic Scenes

    Yi-Hua Huang et al. SC-GS: Sparse-Controlled Gaus- sian Splatting for Editable Dynamic Scenes . 2024. arXiv: 2312.14937 [cs.CV]. URL: https://arxiv.org/ abs/2312.14937

    2024 arXiv

  5. [13]

    GSDeformer: Direct, Real-time and Extensible Cage-based Deformation for 3D Gaussian Splatting

    Jiajun Huang et al. GSDeformer: Direct, Real-time and Extensible Cage-based Deformation for 3D Gaussian Splatting. 2024. arXiv: 2405 . 15491 [cs.CV]. URL: https://arxiv.org/abs/2405.15491

    2024 arXiv

  6. [14]

    Mean Value Coordinates for Closed Triangular Meshes

    Tao Ju, Scott Schaefer, and Joe Warren. “Mean Value Coordinates for Closed Triangular Meshes”. In: Seminal Graphics Papers: Pushing the Boundaries, Volume 2 . 1st ed. New York, NY , USA: Association for Computing Machinery, 2023. ISBN : 9798400708978. URL: https : //doi.org/10...

    2023

  7. [15]

    3D Gaussian Splatting for Real- Time Radiance Field Rendering

    Bernhard Kerbl et al. “3D Gaussian Splatting for Real- Time Radiance Field Rendering”. In: ACM Transactions on Graphics 42.4 (2023)

    2023

  8. [16]

    NeRO: Neural Geometry and BRDF Reconstruction of Reflective Objects from Multiview Images

    Yuan Liu et al. “NeRO: Neural Geometry and BRDF Reconstruction of Reflective Objects from Multiview Images”. In: SIGGRAPH. 2023

    2023

  9. [17]

    Marching cubes: A high resolution 3D surface construction al- gorithm

    William E. Lorensen and Harvey E. Cline. “Marching cubes: A high resolution 3D surface construction al- gorithm”. In: Proceedings of the 14th Annual Confer- ence on Computer Graphics and Interactive Techniques. SIGGRAPH ’87. New York, NY , USA: Association for Computing Machin...

    1987

  10. [18]

    3DGSR: Implicit Surface Recon- struction with 3D Gaussian Splatting

    Xiaoyang Lyu et al. 3DGSR: Implicit Surface Recon- struction with 3D Gaussian Splatting . 2024. arXiv: 2404 . 00409 [cs.CV]. URL: https : / / arxiv. org / abs / 2404.00409

    2024 arXiv

  11. [19]

    NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis

    Ben Mildenhall et al. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis. 2020. arXiv: 2003.08934 [cs.CV]. URL: https://arxiv.org/abs/2003. 08934

    2020 arXiv

  12. [21]

    Instant neural graphics primitives with a multiresolution hash encoding

    Thomas Müller et al. “Instant neural graphics primitives with a multiresolution hash encoding”. In: ACM Trans- actions on Graphics 41.4 (July 2022), pp. 1–15. ISSN : 1557-7368. DOI: 10.1145/3528223.3530127. URL: http: //dx.doi.org/10.1145/3528223.3530127

    2022

  13. [22]

    A survey of the marching cubes algorithm

    Timothy S Newman and Hong Yi. “A survey of the marching cubes algorithm”. In: Computers & Graphics 30.5 (2006), pp. 854–879

    2006

  14. [23]

    A Survey of Structure from Motion

    Onur Ozyesil et al. A Survey of Structure from Motion

  15. [24]

    Nerfies: Deformable Neural Radi- ance Fields

    Keunhong Park et al. Nerfies: Deformable Neural Radi- ance Fields. 2021. arXiv: 2011.12948 [cs.CV]. URL: https://arxiv.org/abs/2011.12948

    2021 arXiv

  16. [25]

    D-NeRF: Neural Radiance Fields for Dynamic Scenes

    Albert Pumarola et al. D-NeRF: Neural Radiance Fields for Dynamic Scenes . 2020. arXiv: 2011.13961 [cs.CV]. URL: https://arxiv.org/abs/2011.13961

    2020 arXiv

  17. [26]

    PermutoSDF: Fast Multi-View Reconstruction with Implicit Surfaces using Permutohedral Lattices

    Radu Alexandru Rosu and Sven Behnke. PermutoSDF: Fast Multi-View Reconstruction with Implicit Surfaces using Permutohedral Lattices. 2023. arXiv: 2211.12562 [cs.CV]. URL: https://arxiv.org/abs/2211.12562

    2023 arXiv

  18. [27]

    Structure-from-Motion Revisited

    Johannes L. Schönberger and Jan-Michael Frahm. “Structure-from-Motion Revisited”. In: 2016 IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR) . 2016, pp. 4104–4113. DOI: 10 . 1109 / CVPR.2016.445

    2016

  19. [28]

    Delicate Textured Mesh Recovery from NeRF via Adaptive Surface Refinement

    Jiaxiang Tang et al. “Delicate Textured Mesh Recovery from NeRF via Adaptive Surface Refinement”. In:arXiv preprint arXiv:2303.02091 (2022)

    2022 arXiv

  20. [29]

    D-MiSo: Editing Dynamic 3D Scenes using Multi-Gaussians Soup

    Joanna Waczy ´nska et al. “D-MiSo: Editing Dynamic 3D Scenes using Multi-Gaussians Soup”. In: arXiv preprint arXiv:2405.14276 (2024)

    2024 arXiv

  21. [30]

    Games: Mesh-based adapt- ing and modification of gaussian splatting

    Joanna Waczy ´nska et al. “Games: Mesh-based adapt- ing and modification of gaussian splatting”. In: arXiv preprint arXiv:2402.01459 (2024)

    2024 arXiv

  22. [31]

    NeuS: Learning Neural Implicit Sur- faces by Volume Rendering for Multi-view Reconstruc- tion

    Peng Wang et al. NeuS: Learning Neural Implicit Sur- faces by Volume Rendering for Multi-view Reconstruc- tion. 2023. arXiv: 2106.10689 [cs.CV]. URL: https: //arxiv.org/abs/2106.10689

    2023 arXiv

  23. [32]

    NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction

    Yiming Wang et al. NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction . 2023. arXiv: 2212.05231 [cs.CV]. URL: https://arxiv.org/ abs/2212.05231

    2023 arXiv

  24. [33]

    PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics

    Tianyi Xie et al. PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics. 2024. arXiv: 2311. 12198 [cs.GR]. URL: https : / / arxiv. org / abs / 2311 . 12198

    2024

  25. [34]

    Deformable 3D Gaussians for High- Fidelity Monocular Dynamic Scene Reconstruction

    Ziyi Yang et al. Deformable 3D Gaussians for High- Fidelity Monocular Dynamic Scene Reconstruction

  26. [35]

    BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks

    Yao Yao et al. “BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks”. In: Com- puter Vision and Pattern Recognition (CVPR) (2020)

    2020

  27. [36]

    BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis

    Lior Yariv et al. BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis . 2023. arXiv: 2302 . 14859 [cs.CV]. URL: https://arxiv.org/abs/2302.14859

    2023 arXiv

  28. [37]

    Plenoxels: Radiance Fields without Neural Networks

    Alex Yu et al. Plenoxels: Radiance Fields without Neural Networks. 2021. arXiv: 2112.05131 [cs.CV]. URL: https://arxiv.org/abs/2112.05131

    2021 arXiv

  29. [2017]

    URL: https://arxiv

    arXiv: 1701.08493 [cs.CV]. URL: https://arxiv. org/abs/1701.08493

    arXiv

  30. [2022]

    URL: https://arxiv

    arXiv: 2203.09517 [cs.CV]. URL: https://arxiv. org/abs/2203.09517

    arXiv

  31. [2023]

    URL: https://arxiv

    arXiv: 2309.13101 [cs.CV]. URL: https://arxiv. org/abs/2309.13101

    arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.