Pith. sign in

REVIEW 3 major objections 4 minor 2 cited by

SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read SurfaceSplat claims that a two-way SDF–Gaussian loop—coarse mesh seeding splats, rendered views refining the mesh—makes sparse-view reconstruction and rendering both improve.

desk verdict Practical two-stage hybrid for sparse-view reconstruction with strong benchmark gains, but the pseudo-view mechanism is not isolated in the ablations and the current evidence is confounded. read the letter →

arxiv 2507.15602 v2 pith:FD2I5I7B submitted 2025-07-21 cs.CV

Zihui Gao , Jia-Wang Bian , Guosheng Lin , Hao Chen , Chunhua Shen This is my paper
classification cs.CV
keywords sparse-viewreconstructionsigneddistancefunction3DGaussiansplattingnovelviewsynthesissurfacepointcloudinitializationpseudo-viewexpansionhybridrepresentation
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

Sparse-view 3D reconstruction fails in two complementary ways: dense signed-distance-function (SDF) models capture global shape but smooth over fine detail, while 3D Gaussian splatting keeps detail but loses global coherence. This paper tries to establish that running the two representations in a loop fixes both weaknesses. A coarse SDF mesh is cleaned and converted into point-cloud initialization for the Gaussians, and the trained Gaussians render extra camera viewpoints that are fed back as training images for a finer SDF pass. On the DTU and MobileBrick benchmarks the paper reports better mesh metrics and novel-view PSNR than the compared SDF-only, Gaussian-only, and joint baselines. The practical stake is improving sparse-view reconstruction with geometric initialization and pseudo-view expansion, without large pretrained generative models.

What carries the argument

The load-bearing object is the two-way data flow between an SDF voxel grid and a set of 3D Gaussians. The SDF-to-Gaussian link is a depth-based surface-point sampler: render depth maps of the coarse mesh at the training views, sample visible pixels, and unproject them to colored 3D points that replace the typical sparse structure-from-motion points as the geometric prior. The Gaussian-to-SDF link is a pose-expansion renderer: cubic-spline interpolation or small position perturbation generates new camera poses, the Gaussians render images from those poses, and the rendered images join the real ones as supervision for a finer SDF pass. Two smaller mechanisms keep the loop stable: a monocular-normal consistency loss that suppresses floaters in the coarse mesh, and a connected-component filter that removes non-dominant mesh pieces before sampling.

What would settle it

Render the expansion views with a Gaussian model initialized from structure-from-motion points instead of from the SDF mesh, keeping everything else identical; if the reported F1 and Chamfer-distance gains over the no-expansion baseline vanish or fall to the level of adding only the monocular normal loss, the pseudo-views are not independent supervision.

Watch

Extended reading notes

Core claim

The central claim, stated on the paper's own terms, is that a single optimization cycle connecting SDF and 3D Gaussian splatting outperforms either representation on its own for sparse inputs. In the SDF-to-Gaussian direction, marching cubes extracts a coarse mesh; connected-component cleaning and a monocular-normal consistency loss remove floating artifacts; depth maps of the cleaned mesh are unprojected to sample visible surface points that initialize the Gaussians, combined with sparse structure-from-motion points for background. In the Gaussian-to-SDF direction, the trained Gaussians render images from perturbed or spline-interpolated camera poses, and those rendered images are added to the original inputs to train a higher-resolution SDF grid. The paper reports that this loop lifts foreground PSNR for rendering and F1/Chamfer distance for geometry on both MobileBrick and DTU, and that a second cycle gives only marginal gains, so the default configuration is one cycle.

Load-bearing premise

The premise that carries the method is that the Gaussian-rendered novel views provide genuinely new, reliable supervision for the SDF, even though the Gaussian model was initialized from the same SDF mesh and trained on the same photographs.

Editorial extensions

If this is right

  • If the central claim holds, sparse-view reconstruction can be improved by better geometric initialization alone, without generative priors: the SDF mesh supplies global structure and the Gaussians supply high-frequency detail.
  • The rendered pseudo-views are the active data-expansion mechanism: adding ten rendered views to ten real inputs increases the reported F1 score on MobileBrick, with gains such as +9.61 at five input views and +2.71 at ten input views, and reduces DTU Chamfer distance.
  • Mesh-sampled initialization benefits 3DGS rendering directly: foreground PSNR on MobileBrick rises from 19.13 with structure-from-motion points to 20.45 with the cleaned, normal-supervised mesh sampling.
  • The loop is representation-agnostic rather than tied to one SDF method: replacing the voxel-grid SDF with another surface reconstruction backbone in the supplementary experiments retains the reported DTU improvement.

Reading between the lines

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

  • If the two-way loop is the active ingredient, the same recipe should transfer to casually captured video with irregular pose coverage, where uniform sparse sampling is unavailable and structure-from-motion points are unreliable; a test would be to run the pipeline on a handheld 360-degree video with pose estimates from a SLAM system.
  • The near-flat gain from a second cycle (F1 68.97 to 69.14, PSNR-F 20.45 to 20.55) suggests that after one expansion the bottleneck shifts from view count to the SDF model's capacity or the quality of the rendered pseudo-views, not to more expansion.
  • Because the Gaussian model is initialized from the same SDF mesh and trained on the same images, the rendered views may be biased toward the initial coarse geometry; we would test this by replacing the expansion images with views rendered by a Gaussian model initialized purely from structure-from-motion points and checking whether the SDF refinement gain survives.
  • The ablation design leaves open how much of the gain comes from the monocular normal loss rather than from the pseudo-views; a controlled experiment adding the normal loss to the no-expansion baseline would separate the two contributions.
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

3 major / 4 minor

Summary. The paper proposes SurfaceSplat, a sparse-view reconstruction pipeline that alternates between SDF-based surface reconstruction and 3D Gaussian Splatting. A coarse SDF mesh (Voxurf) is cleaned, sampled to initialize 3DGS, and then 3DGS renders additional novel-view images that are added to the training set to refine the SDF at higher voxel resolution. Experiments on DTU and MobileBrick report state-of-the-art Chamfer Distance/F1 and PSNR/PSNR-F, with ablations on pose expansion, number of rendered views, and 3DGS initialization.

Significance. If the proposed two-way coupling is the cause of the reported gains, the paper is a solid step toward combining dense and sparse scene representations for sparse-view reconstruction. The manuscript is clearly written, the datasets are standard, the code is promised for release, and the experimental breadth (two datasets, multiple sparsity levels, per-scene tables, and a SparseCraft variant) is a genuine strength. The main reservation is that the ablations do not isolate the pseudo-view-expansion mechanism from the other modifications (normal loss, mesh cleaning, SDF-initialized 3DGS), and the pseudo-views may inherit the coarse SDF geometry rather than provide independent supervision; these points must be addressed before the central claim is fully supported.

major comments (3)
  1. [Sec. 4.3, Tables 3, 5, 9] The ablations do not isolate the contribution of the 3DGS-rendered pseudo-views. In Table 9, the row 'SDF-based method w/o 3DGS' reports F1=62.42, CD=13.3, PSNR=14.34, PSNR-F=18.34, which are exactly the Voxurf numbers in Table 1; in Table 5, the 'Baseline' F1=55.8 is exactly the Voxurf value for aston in Table 7; and in Table 3, the 'Baseline' on DTU scan69 is CD=2.940, again the Voxurf value. Thus these baselines are plain Voxurf, not SurfaceSplat with only the pseudo-view expansion removed. The reported 3DGS-for-SDF gain (about 5 F1 points in Table 5 and 6.5 F1 points in Table 9) could therefore be caused entirely by the monocular normal loss (Eq. 4), the mesh cleaning (Sec. 3.2), the SDF-mesh initialization of 3DGS, or some combination, rather than by the rendered pseudo-views that are the mechanism claimed in Sec. 3.3 and the abstract. I ask for an ablation that removes only the pseudo-views from the pipeline (keeping the normal loss, cleaning, and 3DGS stage) and for the baseline rows to be relabeled as Voxurf if that is what they are.
  2. [Sec. 3.3, Eq. (7)] The pseudo-views are rendered by a 3DGS model initialized from the same coarse SDF mesh and trained on the same input images, so the supervision they provide to the SDF may be largely circular: the SDF is refined on images that already encode its own coarse geometry. Equation (7) is a deterministic splatting of G with no confidence or multi-view-consistency filter before {I_new} is added to the SDF training set. I do not claim the mechanism is certainly circular, but the present experiments do not rule it out. A concrete and feasible test would be to compare the current pseudo-view expansion against expansion using views rendered from a 3DGS initialized only from COLMAP points, and to measure whether the refinement improves on held-out real views; alternatively, report the photometric or geometric consistency error between rendered pseudo-views and the original inputs for the added viewpoints.
  3. [Sec. 3.2, Eq. (4) vs. Sec. 3.3] The improvements reported in Tables 4 and 5 are consistent with an alternative explanation: the monocular normal loss alone, plus the largest-component mesh cleaning, may be responsible for most of the gains, and the pseudo-view expansion may add little once those components are present. Table 4 shows that 3DGS rendering improves from 19.13 to 20.45 with 'normal and mesh clean', but it does not include a row with the normal loss alone or with mesh cleaning alone, so the individual contributions are not separable. Because the normal loss is already a strong prior in sparse-view SDF reconstruction (as in MonoSDF), the paper should show a row of SurfaceSplat without the rendered views but with all other components, or equivalently a row with pseudo-views but without the normal loss, to attribute the Table 3 and Table 9 gains specifically to the 3DGS-for-SDF coupling.
minor comments (4)
  1. [Sec. 4.3, Table 5 caption] The caption contains a typo: 'pose expansion strategies for in MobileBrick' should read 'pose expansion strategies on MobileBrick'.
  2. [Sec. 2.3 heading] The heading 'Combing 3DGS and SDF' should be 'Combining 3DGS and SDF'.
  3. [Sec. 3.2 and Sec. 4.1 references] The COLMAP reference is given as [33] in Sec. 3.2 and [31] in Sec. 4.1; the citation index should be consistent.
  4. [Sec. 3.4, Eq. (10)] The symbol R is used for the rendering operator in Eq. (10) and for camera rotation matrices elsewhere; please use distinct notation to avoid ambiguity.

Circularity Check

2 steps flagged · score 6.0 of 10

Pseudo-view refinement is a self-referential loop and its ablations compare against plain Voxurf/GOF, so the central 3DGS-for-SDF gain is not isolated.

  1. self definitional [Section 3.4, Eqs. (10)-(12); see also Section 3.3 Eq. (7)]
    "Rendering Step: We optimize a 3DGS model for rendering novel view images, which is initialized by sampling points from the current coarse mesh M(n)c, represented by: I(n) = R(M(n)c) (10). ... Meshing Step: We refine the current mesh by fine-tuning it using both the newly rendered images and the original input images: M(n)f = O(M(n)c, I(n)) (11)."

    The images used to refine the SDF mesh are renderings of a 3DGS initialized from that same mesh (surface points P are sampled from M(n)c via Eq. 6) and trained on the same input views. Hence I(n) is a re-projection of the mesh's own geometry; the fine-stage SDF is optimized against images that encode its own coarse state. No confidence or consistency filter is applied before adding the rendered images to the training set. The 'new' supervision is therefore constructed from the quantity it is supposed to correct, so the refinement can reinforce rather than correct coarse-SDF errors. This is a self-training loop, not independent evidence.

  2. fitted input called prediction [Section 4.3, Table 3 and Table 5; Table 9]
    "Table 3 compares our method with a pure SDF-based reconstruction baseline at different sparsity levels ... Baseline 55.8 49.9 8.7 (Table 5)"

    The baseline is not the proposed method with only the pseudo-view module removed: the value 55.8 in Table 5 is exactly Voxurf's Aston F1 score in Table 7, and Table 9's 'SDF-based method w/o 3DGS' row (62.42, 13.3, 14.34, 18.34) is Voxurf's Table 1 row; '3DGS-based method w/o SDF' equals GOF. Thus the reported improvement bundles the monocular normal loss (Eq. 4), mesh cleaning (Eq. 5), and mesh-point 3DGS initialization together with the pseudo-view expansion. The causal contribution of the rendered pseudo-views to the headline SDF gain is not isolated, so the ablation does not validate the '3DGS for Enhanced SDF' mechanism it is cited for.

full rationale

The paper's benchmark comparisons are external and reproducible in principle, and there is no self-citation chain that forces the result. However, the central claimed mechanism—3DGS-rendered pseudo-views refining the SDF—is supported only by (i) an optimization loop in which the supervising images are rendered from a 3DGS initialized from the very SDF mesh being refined (Eqs. 6, 7, 10, 11), and (ii) ablations whose 'Baseline' rows are unmodified Voxurf/GOF numbers identical to the comparison tables. The first is a self-training loop that can reinforce the coarse SDF rather than add independent detail; the second means the measured gain cannot be attributed to the pseudo-view module as opposed to the added normal loss, mesh cleaning, or mesh-based initialization. These are partial circularity and confounding issues in the evidence chain, not a formal tautology, so the score is 6 rather than higher.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

The central claim rests on several hand-chosen hyperparameters (normal loss weight, number of rendered views, sampling density, grid resolutions) and on the assumption that 3DGS-rendered pseudo-views provide reliable supervision. The only ad hoc entity-like assumption is the reliability of the rendered views; no new physical or conceptual entities are introduced.

free parameters (7)
  • normal_loss_weight = 0.05
    Weight for the monocular normal consistency loss (Eq. 4). Set by hand, not ablated.
  • num_rendered_views = 10
    Number of pseudo-views rendered by 3DGS for SDF augmentation. Chosen based on Fig. 7, not derived.
  • sampled_points_per_depth_map = 5000
    Number of points sampled per training view depth map, used for 3DGS initialization.
  • total_sampled_points = 50000
    Fused point count after combining sampled depth points, used to initialize 3DGS.
  • coarse_voxel_resolution = 96
    Voxel grid resolution for the coarse SDF stage.
  • fine_voxel_resolution = 256
    Voxel grid resolution for the fine SDF refinement stage.
  • 3dgs_iterations = 7000
    Training iterations for 3DGS under sparse views; convergence was checked informally.
assumptions (4)
  • standard math SDF volume rendering (Eqs. 1-2) provides differentiable photometric supervision
    The paper adopts NeuS and Voxurf rendering equations without proof, treating them as established background.
  • domain assumption Metric3D monocular normals are accurate enough to supervise SDF geometry
    The normal loss (Eq. 4) uses predictions from Metric3D as pseudo-ground-truth normals; the paper does not quantify Metric3D errors on these scenes.
  • domain assumption COLMAP provides accurate camera poses and sparse points
    The method relies on COLMAP poses and combines COLMAP points with mesh-sampled points, assuming the poses are sufficiently accurate.
  • ad hoc to paper Rendered pseudo-views are reliable supervision for SDF refinement
    The entire 3DGS-to-SDF enhancement in Sec 3.3 assumes that images rendered by 3DGS from novel poses are accurate enough to train the SDF; this is the core assumption of the feedback loop.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting." pith.science (2026). https://pith.science/paper/FD2I5I7B

@misc{pith2026250715602,
  author       = {Pith},
  title        = {Pith review of: SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FD2I5I7B}},
  note         = {Machine review of arXiv:2507.15602}
}
read the original abstract

Surface reconstruction and novel view rendering from sparse-view images are challenging. Signed Distance Function (SDF)-based methods struggle with fine details, while 3D Gaussian Splatting (3DGS)-based approaches lack global geometry coherence. We propose a novel hybrid method that combines the strengths of both approaches: SDF captures coarse geometry to enhance 3DGS-based rendering, while newly rendered images from 3DGS refine the details of SDF for accurate surface reconstruction. As a result, our method surpasses state-of-the-art approaches in surface reconstruction and novel view synthesis on the DTU and MobileBrick datasets. Code will be released at https://github.com/aim-uofa/SurfaceSplat.

Figures

Figures reproduced from arXiv: 2507.15602 by the authors.

Figure 1
Figure 1. Sparse view reconstruction and rendering comparison. Left: Qualitative results from 10 images evenly sampled from a casually captured 360-degree video. Right: Quantitative analysis of 5, 10, and 20 input views, averaged across the selected 9 MobileBrick test scenes. 3DGS-based methods (e.g., GOF) achieve superior novel view rendering than SDF-based methods (e.g., Voxurf) due to their sparse representations, which ca… view at source ↗
Figure 2
Figure 2. Overview of the proposed SurfaceSplat. (A) We reconstruct a coarse mesh using an SDF-based representation. (B) Point clouds [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Visualization of our mesh reconstruction, cleaning, and [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Top-view visualization of pose expansion strategies. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 7
Figure 7. Figure 7: econstruction quality with varying numbers of 3DGS [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 5
Figure 5. Figure 5: Qualitative mesh reconstruction comparisons on MobileBrick. See more visual results in supplementary material. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Qualitative novel view synthesis comparisons on MobileBrick. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 8
Figure 8. Figure 8: Ablation study on mesh-based sampling for enhancing 3DGS rendering. We report foreground PSNR here. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 10
Figure 10. Figure 10: Qualitative comparison of 3-view mesh reconstruction [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: DTU novel view synthesis comparison. C.3. MobileBrick rendering results [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]
Figure 9
Figure 9. Figure 9: Visualization of newly rendered images with differ [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 12
Figure 12. Figure 12: More qualitative mesh reconstruction results on MobileBrick and DTU. [PITH_FULL_IMAGE:figures/full_fig_p015_12.png]
Figure 13
Figure 13. Figure 13: More qualitative novel view rendering results on MobileBrick. [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. TriSplat: Simulation-Ready Feed-Forward 3D Scene Reconstruction

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    TriSplat predicts oriented triangle primitives from images in one forward pass to produce simulation-ready 3D meshes with competitive rendering quality.

  2. Feed-Forward 3D Scene Modeling: A Problem-Driven Perspective

    cs.CV 2026-04 unverdicted novelty 6.0 of 10

    The paper proposes a problem-driven taxonomy for feed-forward 3D scene modeling that groups methods by five core challenges: feature enhancement, geometry awareness, model efficiency, augmentation strategies, and temp...

Reference graph

Works this paper leans on

59 extracted references · 47 canonical work pages · cited by 2 Pith papers

  1. [1]

    360-gs: Layout-guided panoramic gaus- sian splatting for indoor roaming

    Jiayang Bai, Letian Huang, Jie Guo, Wen Gong, Yuanqi Li, and Yanwen Guo. 360-gs: Layout-guided panoramic gaus- sian splatting for indoor roaming. In International Confer- ence on 3D Vision 2025. 2

    work page 2025

  2. [2]

    Mip-nerf: A multiscale representation for anti-aliasing neu- ral radiance fields

    Jonathan T Barron, Ben Mildenhall, Matthew Tancik, Peter Hedman, Ricardo Martin-Brualla, and Pratul P Srinivasan. Mip-nerf: A multiscale representation for anti-aliasing neu- ral radiance fields. In Int. Conf. Comput. Vis., pages 5855– 5864, 2021. 2

    work page 2021

  3. [3]

    Mip-nerf 360: Unbounded anti-aliased neural radiance fields

    Jonathan T Barron, Ben Mildenhall, Dor Verbin, Pratul P Srinivasan, and Peter Hedman. Mip-nerf 360: Unbounded anti-aliased neural radiance fields. In IEEE Conf. Comput. Vis. Pattern Recog., pages 5470–5479, 2022. 2

    work page 2022

  4. [4]

    Porf: Pose residual field for accurate neural sur- face reconstruction

    Jia-Wang Bian, Wenjing Bian, Victor Adrian Prisacariu, and Philip Torr. Porf: Pose residual field for accurate neural sur- face reconstruction. In ICLR, 2024. 2

    work page 2024

  5. [5]

    Nope-nerf: Optimising neural radi- ance field with no pose prior

    Wenjing Bian, Zirui Wang, Kejie Li, Jiawang Bian, and Vic- tor Adrian Prisacariu. Nope-nerf: Optimising neural radi- ance field with no pose prior. 2023. 2

    work page 2023

  6. [6]

    Mvsnerf: Fast general- izable radiance field reconstruction from multi-view stereo

    Anpei Chen, Zexiang Xu, Fuqiang Zhao, Xiaoshuai Zhang, Fanbo Xiang, Jingyi Yu, and Hao Su. Mvsnerf: Fast general- izable radiance field reconstruction from multi-view stereo. In Int. Conf. Comput. Vis., pages 14124–14133, 2021. 1, 2

    work page 2021

  7. [7]

    Tensorf: Tensorial radiance fields

    Anpei Chen, Zexiang Xu, Andreas Geiger, Jingyi Yu, and Hao Su. Tensorf: Tensorial radiance fields. In Eur. Conf. Comput. Vis., pages 333–350. Springer, 2022. 2

    work page 2022

  8. [8]

    Pgsr: Planar-based gaussian splatting for ef- ficient and high-fidelity surface reconstruction

    Danpeng Chen, Hai Li, Weicai Ye, Yifan Wang, Weijian Xie, Shangjin Zhai, Nan Wang, Haomin Liu, Hujun Bao, and Guofeng Zhang. Pgsr: Planar-based gaussian splatting for ef- ficient and high-fidelity surface reconstruction. IEEE Trans- actions on Visualization and Computer Graphics, 2024. 2

    work page 2024

Show all 59 references
  1. [9]

    Neusg: Neural im- plicit surface reconstruction with 3d gaussian splatting guid- ance

    Hanlin Chen, Chen Li, and Gim Hee Lee. Neusg: Neural im- plicit surface reconstruction with 3d gaussian splatting guid- ance. arXiv preprint arXiv:2312.00846, 2023. 2

    2023 arXiv

  2. [10]

    Meshanything: Artist-created mesh generation with au- toregressive transformers

    Yiwen Chen, Tong He, Di Huang, Weicai Ye, Sijin Chen, Ji- axiang Tang, Xin Chen, Zhongang Cai, Lei Yang, Gang Yu, et al. Meshanything: Artist-created mesh generation with au- toregressive transformers. arXiv preprint arXiv:2406.10163,

    arXiv

  3. [11]

    A practical guide to splines

    C De Boor. A practical guide to splines. Springer-Verlag google schola, 2:4135–4195, 1978. 5

    1978

  4. [12]

    Plenoxels: Radiance fields without neural networks

    Sara Fridovich-Keil, Alex Yu, Matthew Tancik, Qinhong Chen, Benjamin Recht, and Angjoo Kanazawa. Plenoxels: Radiance fields without neural networks. In IEEE Conf. Comput. Vis. Pattern Recog., pages 5501–5510, 2022. 2

    2022

  5. [13]

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

    Antoine Gu ´edon and Vincent Lepetit. Sugar: Surface- aligned gaussian splatting for efficient 3d mesh reconstruc- tion and high-quality mesh rendering. In IEEE Conf. Com- put. Vis. Pattern Recog., pages 5354–5363, 2024. 2

    2024

  6. [14]

    From nerfs to gaussian splats, and back

    Siming He, Zach Osman, and Pratik Chaudhari. From nerfs to gaussian splats, and back. arXiv preprint arXiv:2405.09717, 2024. 2

    2024 arXiv

  7. [15]

    2d gaussian splatting for geometrically accu- rate radiance fields

    Binbin Huang, Zehao Yu, Anpei Chen, Andreas Geiger, and Shenghua Gao. 2d gaussian splatting for geometrically accu- rate radiance fields. In SIGGRAPH 2024 Conference Papers. Association for Computing Machinery, 2024. 1, 2, 6, 7

    2024

  8. [16]

    Large scale multi-view stereopsis evalu- ation

    Rasmus Jensen, Anders Dahl, George V ogiatzis, Engil Tola, and Henrik Aanæs. Large scale multi-view stereopsis evalu- ation. In IEEE Conf. Comput. Vis. Pattern Recog., 2014. 2, 5, 3

    2014

  9. [17]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 2023. 1, 2, 3, 6, 7

    2023

  10. [18]

    Integrating meshes and 3d gaussians for indoor scene reconstruction with sam mask guidance

    Jiyeop Kim and Jongwoo Lim. Integrating meshes and 3d gaussians for indoor scene reconstruction with sam mask guidance. arXiv preprint arXiv:2407.16173, 2024. 2

    2024 arXiv

  11. [19]

    Mobilebrick: Building lego for 3d reconstruction on mobile devices

    Kejie Li, Jia-Wang Bian, Robert Castle, Philip HS Torr, and Victor Adrian Prisacariu. Mobilebrick: Building lego for 3d reconstruction on mobile devices. In IEEE Conf. Comput. Vis. Pattern Recog., pages 4892–4901, 2023. 2, 5, 1

    2023

  12. [20]

    Neuralangelo: High-fidelity neural surface reconstruction

    Zhaoshuo Li, Thomas M ¨uller, Alex Evans, Russell H Tay- lor, Mathias Unberath, Ming-Yu Liu, and Chen-Hsuan Lin. Neuralangelo: High-fidelity neural surface reconstruction. In IEEE Conf. Comput. Vis. Pattern Recog., pages 8456–8465,

  13. [21]

    Retr: Modeling rendering via transformer for generalizable neural surface re- construction

    Yixun Liang, Hao He, and Yingcong Chen. Retr: Modeling rendering via transformer for generalizable neural surface re- construction. Advances in Neural Information Processing Systems, 36:62332–62351, 2023. 3

    2023

  14. [22]

    Sparseneus: Fast generalizable neural sur- face reconstruction from sparse views

    Xiaoxiao Long, Cheng Lin, Peng Wang, Taku Komura, and Wenping Wang. Sparseneus: Fast generalizable neural sur- face reconstruction from sparse views. InEur. Conf. Comput. Vis., pages 210–227. Springer, 2022. 1, 2, 6, 7, 3

    2022

  15. [23]

    Marching cubes: A high resolution 3d surface construction algorithm

    William E Lorensen and Harvey E Cline. Marching cubes: A high resolution 3d surface construction algorithm. InSem- inal graphics: pioneering efforts that shaped the field, pages 347–353, 1998. 2

    1998

  16. [24]

    Scaffold-gs: Structured 3d gaussians for view-adaptive rendering

    Tao Lu, Mulin Yu, Linning Xu, Yuanbo Xiangli, Limin Wang, Dahua Lin, and Bo Dai. Scaffold-gs: Structured 3d gaussians for view-adaptive rendering. In IEEE Conf. Com- put. Vis. Pattern Recog., pages 20654–20664, 2024. 2

    2024

  17. [25]

    3dgsr: Implicit surface reconstruction with 3d gaussian splat- ting

    Xiaoyang Lyu, Yang-Tian Sun, Yi-Hua Huang, Xiuzhe Wu, Ziyi Yang, Yilun Chen, Jiangmiao Pang, and Xiaojuan Qi. 3dgsr: Implicit surface reconstruction with 3d gaussian splat- ting. ACM Transactions on Graphics (TOG) , 43(6):1–12,

  18. [26]

    Srinivasan, Matthew Tancik, Jonathan T

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis. In Eur. Conf. Comput. Vis., 2020. 1, 2

    2020

  19. [27]

    Instant neural graphics primitives with a mul- tiresolution hash encoding

    Thomas M ¨uller, Alex Evans, Christoph Schied, and Alexan- der Keller. Instant neural graphics primitives with a mul- tiresolution hash encoding. ACM Trans. Graph., 41(4):1–15,

  20. [28]

    Constructing signed distance functions

    Stanley Osher, Ronald Fedkiw, Stanley Osher, and Ronald Fedkiw. Constructing signed distance functions. Level set methods and dynamic implicit surfaces , pages 63–74, 2003. 2

    2003

  21. [29]

    Deepsdf: Learning con- tinuous signed distance functions for shape representation

    Jeong Joon Park, Peter Florence, Julian Straub, Richard Newcombe, and Steven Lovegrove. Deepsdf: Learning con- tinuous signed distance functions for shape representation. In IEEE Conf. Comput. Vis. Pattern Recog., pages 165–174,

  22. [30]

    V olrecon: V olume rendering of signed ray distance functions for generalizable multi-view recon- struction

    Yufan Ren, Fangjinhua Wang, Tong Zhang, Marc Pollefeys, and Sabine S¨usstrunk. V olrecon: V olume rendering of signed ray distance functions for generalizable multi-view recon- struction. In IEEE Conf. Comput. Vis. Pattern Recog., pages 16685–16695, 2023. 1, 2, 3

    2023

  23. [31]

    Structure-from-motion revisited

    Johannes Lutz Sch ¨onberger and Jan-Michael Frahm. Structure-from-motion revisited. In IEEE Conf. Comput. Vis. Pattern Recog., 2016. 6

    2016

  24. [32]

    Pixelwise view selection for unstructured multi-view stereo

    Johannes L Sch ¨onberger, Enliang Zheng, Jan-Michael Frahm, and Marc Pollefeys. Pixelwise view selection for unstructured multi-view stereo. In Eur. Conf. Comput. Vis., pages 501–518. Springer, 2016. 1

    2016

  25. [33]

    Photo tourism: exploring photo collections in 3d

    Noah Snavely, Steven M Seitz, and Richard Szeliski. Photo tourism: exploring photo collections in 3d. In ACM Sig- graph, pages 835–846, 2006. 3

    2006

  26. [34]

    Sa- gs: Scale-adaptive gaussian splatting for training-free anti- aliasing

    Xiaowei Song, Jv Zheng, Shiran Yuan, Huan-ang Gao, Jing- wei Zhao, Xiang He, Weihao Gu, and Hao Zhao. Sa- gs: Scale-adaptive gaussian splatting for training-free anti- aliasing. arXiv preprint arXiv:2403.19615, 2024. 2

    2024 arXiv

  27. [35]

    Direct voxel grid optimization: Super-fast convergence for radiance fields reconstruction

    Cheng Sun, Min Sun, and Hwann-Tzong Chen. Direct voxel grid optimization: Super-fast convergence for radiance fields reconstruction. In IEEE Conf. Comput. Vis. Pattern Recog., pages 5459–5469, 2022. 2

    2022

  28. [36]

    3dgs-hd: Elimination of unreal- istic artifacts in 3d gaussian splatting

    Hao Sun, Junping Qin, Lei Wang, Kai Yan, Zheng Liu, Xin- glong Jia, and Xiaole Shi. 3dgs-hd: Elimination of unreal- istic artifacts in 3d gaussian splatting. In 2024 6th Interna- tional Conference on Data-driven Optimization of Complex Systems (DOCS), pages 696–702. IEEE, 2024. 2, 1

    2024

  29. [37]

    Dn-splatter: Depth and normal priors for gaussian splatting and meshing

    Matias Turkulainen, Xuqian Ren, Iaroslav Melekhov, Otto Seiskari, Esa Rahtu, and Juho Kannala. Dn-splatter: Depth and normal priors for gaussian splatting and meshing. In 2025 IEEE/CVF Winter Conference on Applications of Com- puter Vision (WACV), pages 2421–2431. IEEE, 2025. 2

    2025

  30. [38]

    Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction

    Peng Wang, Lingjie Liu, Yuan Liu, Christian Theobalt, Taku Komura, and Wenping Wang. Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction. In Adv. Neural Inform. Process. Syst., 2021. 1, 2, 5

    2021

  31. [39]

    Enhancement of 3d gaussian splatting using raw mesh for photorealistic recreation of ar- chitectures

    Ruizhe Wang, Chunliang Hua, Tomakayev Shingys, Mengyuan Niu, Qingxin Yang, Lizhong Gao, Yi Zheng, Jun- yan Yang, and Qiao Wang. Enhancement of 3d gaussian splatting using raw mesh for photorealistic recreation of ar- chitectures. arXiv preprint arXiv:2407.15435, 2024. 2

    2024 arXiv

  32. [40]

    Marching cubes: A high resolution 3d surface construction algorithm

    LORENSEN WE. Marching cubes: A high resolution 3d surface construction algorithm. Computer graphics, 21(1): 7–12, 1987. 4

    1987

  33. [41]

    Meshlrm: Large reconstruction model for high- quality meshes

    Xinyue Wei, Kai Zhang, Sai Bi, Hao Tan, Fujun Luan, Valentin Deschaintre, Kalyan Sunkavalli, Hao Su, and Zex- iang Xu. Meshlrm: Large reconstruction model for high- quality meshes. arXiv preprint arXiv:2404.12385, 2024. 1, 2

    2024 arXiv

  34. [42]

    Direct3d: Scal- able image-to-3d generation via 3d latent diffusion trans- former

    Shuang Wu, Youtian Lin, Feihu Zhang, Yifei Zeng, Jingxi Xu, Philip Torr, Xun Cao, and Yao Yao. Direct3d: Scal- able image-to-3d generation via 3d latent diffusion trans- former. Advances in Neural Information Processing Systems, 37:121859–121881, 2024. 1, 2

    2024

  35. [43]

    V oxurf: V oxel-based efficient and accurate neural surface reconstruction

    Tong Wu, Jiaqi Wang, Xingang Pan, Xudong Xu, Christian Theobalt, Ziwei Liu, and Dahua Lin. V oxurf: V oxel-based efficient and accurate neural surface reconstruction. In Int. Conf. Learn. Represent., 2023. 1, 3, 4, 5, 6, 7, 2

    2023

  36. [44]

    Sparsegs: Real- time 360° sparse view synthesis using gaussian splatting

    Haolin Xiong, Sairisheek Muttukuru, Rishi Upadhyay, Pradyumna Chari, and Achuta Kadambi. Sparsegs: Real- time 360° sparse view synthesis using gaussian splatting. Arxiv, 2023. 6

    2023

  37. [45]

    Instantmesh: Efficient 3d mesh generation from a single image with sparse-view large reconstruction models

    Jiale Xu, Weihao Cheng, Yiming Gao, Xintao Wang, Shenghua Gao, and Ying Shan. Instantmesh: Efficient 3d mesh generation from a single image with sparse-view large reconstruction models. arXiv preprint arXiv:2404.07191 ,

    arXiv

  38. [46]

    Spectrally pruned gaussian fields with neural compensation

    Runyi Yang, Zhenxin Zhu, Zhou Jiang, Baijun Ye, Xiaoxue Chen, Yifei Zhang, Yuantao Chen, Jian Zhao, and Hao Zhao. Spectrally pruned gaussian fields with neural compensation. arXiv preprint arXiv:2405.00676, 2024. 2

    2024 arXiv

  39. [47]

    Deformable 3d gaussians for high- fidelity monocular dynamic scene reconstruction

    Ziyi Yang, Xinyu Gao, Wen Zhou, Shaohui Jiao, Yuqing Zhang, and Xiaogang Jin. Deformable 3d gaussians for high- fidelity monocular dynamic scene reconstruction. In IEEE Conf. Comput. Vis. Pattern Recog. , pages 20331–20341,

  40. [48]

    Blendedmvs: A large- scale dataset for generalized multi-view stereo networks

    Yao Yao, Zixin Luo, Shiwei Li, Jingyang Zhang, Yufan Ren, Lei Zhou, Tian Fang, and Long Quan. Blendedmvs: A large- scale dataset for generalized multi-view stereo networks. In IEEE Conf. Comput. Vis. Pattern Recog., pages 1790–1799,

  41. [49]

    V ol- ume rendering of neural implicit surfaces

    Lior Yariv, Jiatao Gu, Yoni Kasten, and Yaron Lipman. V ol- ume rendering of neural implicit surfaces. Adv. Neural In- form. Process. Syst., 34:4805–4815, 2021. 2

    2021

  42. [50]

    gsplat: An open-source library for gaussian splatting

    Vickie Ye, Ruilong Li, Justin Kerr, Matias Turkulainen, Brent Yi, Zhuoyang Pan, Otto Seiskari, Jianbo Ye, Jeffrey Hu, Matthew Tancik, et al. gsplat: An open-source library for gaussian splatting. Journal of Machine Learning Research , 26(34):1–17, 2025. 2

    2025

  43. [51]

    Metric3d: Towards zero-shot metric 3d prediction from a single image

    Wei Yin, Chi Zhang, Hao Chen, Zhipeng Cai, Gang Yu, Kaixuan Wang, Xiaozhi Chen, and Chunhua Shen. Metric3d: Towards zero-shot metric 3d prediction from a single image. In Int. Conf. Comput. Vis., pages 9043–9053, 2023. 4

    2023

  44. [52]

    Spar- secraft: Few-shot neural reconstruction through stereopsis guided geometric linearization

    Mae Younes, Amine Ouasfi, and Adnane Boukhayma. Spar- secraft: Few-shot neural reconstruction through stereopsis guided geometric linearization. In European Conference on Computer Vision, pages 37–56. Springer, 2024. 3

    2024

  45. [53]

    pixelnerf: Neural radiance fields from one or few images

    Alex Yu, Vickie Ye, Matthew Tancik, and Angjoo Kanazawa. pixelnerf: Neural radiance fields from one or few images. In IEEE Conf. Comput. Vis. Pattern Recog., pages 4578–4587,

  46. [54]

    Gsdf: 3dgs meets sdf for improved neural ren- dering and reconstruction

    Mulin Yu, Tao Lu, Linning Xu, Lihan Jiang, Yuanbo Xiangli, and Bo Dai. Gsdf: 3dgs meets sdf for improved neural ren- dering and reconstruction. Advances in Neural Information Processing Systems, 37:129507–129530, 2024. 2, 6, 7

    2024

  47. [55]

    Monosdf: Exploring monocu- lar geometric cues for neural implicit surface reconstruction

    Zehao Yu, Songyou Peng, Michael Niemeyer, Torsten Sat- tler, and Andreas Geiger. Monosdf: Exploring monocu- lar geometric cues for neural implicit surface reconstruction. Adv. Neural Inform. Process. Syst., 2022. 6, 1, 2, 3

    2022

  48. [56]

    Mip-splatting: Alias-free 3d gaussian splat- ting

    Zehao Yu, Anpei Chen, Binbin Huang, Torsten Sattler, and Andreas Geiger. Mip-splatting: Alias-free 3d gaussian splat- ting. In IEEE Conf. Comput. Vis. Pattern Recog. , pages 19447–19456, 2024. 2

    2024

  49. [57]

    Gaussian opacity fields: Efficient adaptive surface reconstruction in unbounded scenes

    Zehao Yu, Torsten Sattler, and Andreas Geiger. Gaussian opacity fields: Efficient adaptive surface reconstruction in unbounded scenes. ACM Trans. Graph., 2024. 1, 2, 6, 7

    2024

  50. [58]

    Rade-gs: Rasterizing depth in gaussian splatting

    Baowen Zhang, Chuan Fang, Rakesh Shrestha, Yixun Liang, Xiaoxiao Long, and Ping Tan. Rade-gs: Rasterizing depth in gaussian splatting. arXiv preprint arXiv:2406.01467, 2024. 2

    2024 arXiv

  51. [59]

    Nerf++: Analyzing and improving neural radiance fields

    Kai Zhang, Gernot Riegler, Noah Snavely, and Vladlen Koltun. Nerf++: Analyzing and improving neural radiance fields. arXiv preprint arXiv:2010.07492, 2020. 2 SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting Supplementary Material A. Implementation Details...

    2010 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.