Pith. sign in

REVIEW 3 major objections 5 minor 43 references

Correlating neighboring surface points with a convolutional neural shader yields sharper multi-view 3D meshes than single-point neural renderers.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-31 16:50 UTC pith:DCGLZAOF

load-bearing objection Useful mesh-based multi-view recon paper: conv shading + rendering-coord displacement beats Neuralangelo on DTU/THuman hard cases; novelty is incremental on NDS but the empirics hold inside the stated scope. the 3 major comments →

arxiv 2607.28132 v1 pith:DCGLZAOF submitted 2026-07-30 cs.CV

Convolutional Neural Shading for High-Quality 3D Reconstruction from Multi-View Images

Juheon Hwang , Taewan Kim , Heeseok Oh , Jiwoo Kang This is my paper
classification cs.CV
keywords multi-view 3D reconstructionconvolutional neural shadingmesh optimizationneural deferred shadingfine-detail displacementdifferentiable rasterizationtextureless surfaces
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Current neural 3D reconstruction methods color a surface from the geometry of one point at a time, so they miss local context and produce blurry or broken geometry in dark, textureless, or boundary regions. This paper replaces that single-point MLP shader with a convolutional neural shader that reads neighboring positions and normals in image space, plus a small displacement network that adds fine offsets to the rasterized mesh. The combination recovers high-frequency wrinkles and clean silhouettes while still using an ordinary triangle mesh, so training stays cheaper than full volumetric NeRF-style methods. On standard benchmarks the approach cuts Chamfer distance below the previous best numbers and visibly improves faces, clothing folds, and sparse-view cases. A sympathetic reader cares because the fix is architectural rather than data-hungry: once the shader can see local neighborhoods, the same multi-view images suddenly support much more accurate geometry.

Core claim

The central claim is that the lack of local geometric detail in recent neural multi-view reconstruction stems directly from single-point shading, and that replacing the MLP shader with convolutional layers over neighboring rasterized positions and normals, together with a learned fine-detail displacement field, produces substantially more accurate meshes and renderings, especially on dark, textureless, and boundary regions.

What carries the argument

Convolutional Neural Shading (CNS): a fine-detail displacement network that predicts per-pixel offsets to position and normal after rasterization, followed by a convolutional neural shader that predicts color from the refined local neighborhood rather than from an isolated point.

Load-bearing premise

An initial coarse mesh inside a bounded visual hull, refined only by progressive subdivision and mask supervision, is already close enough that the convolutional gradients recover true surface detail instead of merely painting view-consistent appearance.

What would settle it

Retrain the identical pipeline on the 15 DTU scenes after swapping the convolutional shader for a standard point-wise MLP and removing the displacement network; if mean Chamfer distance does not rise well above the reported 0.49 and visual wrinkles disappear, the claimed benefit of spatial correlation is false.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Mesh-based multi-view pipelines can match or beat implicit-surface accuracy without volumetric sampling costs.
  • Dark and textureless objects become routinely reconstructible once the shader sees local neighborhoods.
  • Sparse-view reconstruction improves because image-boundary discontinuities are regularized by the displacement network.
  • Existing mesh optimizers can adopt the same convolutional shader module as a drop-in replacement for MLP shading.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same neighborhood-correlation idea should transfer to dynamic or deformable meshes if the displacement network is conditioned on time or pose.
  • Failure modes on unbounded or heavily occluded scenes suggest the next natural extension is a hierarchical or visibility-aware rasterizer rather than a return to full volumes.
  • Because the method already separates diffuse and specular heads, it may serve as a lightweight inverse-rendering front-end for material editing once lighting is explicitly parameterized.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes Convolutional Neural Shading (CNS), a mesh-based multi-view reconstruction pipeline that replaces pointwise MLP shading with a convolutional neural shader operating on fine-detail positions and normals, plus a fine-detail displacement network that predicts per-pixel offsets in rendering coordinates. Starting from a coarse mesh (primitive or shape-from-silhouette), the method jointly optimizes mesh vertices and the two networks under mesh regularizers and image losses (RGB, mask, SSIM, LPIPS), with progressive subdivision. Experiments on DTU, BlendedMVS, THuman (dark/textureless subset), and a custom capture report lower Chamfer distances than IDR, NeuS, NeuralUDF, Neuralangelo, SuGaR and others (DTU mean 0.49 vs Neuralangelo 0.61; THuman dark 0.41 vs 0.72), with supporting qualitative results, error maps, sparse-view tests, and ablations removing each module.

Significance. If the gains hold under full reproducibility, the work is a useful practical advance for mesh-based neural reconstruction: it shows that spatial context in the shader and explicit displacement in image space measurably improve geometry in the regimes where single-point MLP shaders fail (dark/textureless surfaces, image boundaries, sparse views). The empirical package is reasonably strong—public GT Chamfer on 15 DTU scenes, a dedicated dark-object split, error-map visualization, and within-pipeline ablations (Table 3). The limitations section candidly restricts the claim to bounded, lightly occluded scenes. The contribution is incremental rather than foundational, but the concrete geometry improvements and the clear mesh-vs-implicit ablation (Table 4) make it of interest to the multi-view reconstruction community.

major comments (3)
  1. [§3.2, §4.1] §3.2 (Eq. 3) and §4.1 leave the convolutional neural shader and the fine-detail displacement network architecturally underspecified. The text states only that the shader comprises separate diffuse/specular networks, that normals and view direction are concatenated “in the middle of the layers,” and that positional encoding uses four levels; no layer counts, channel widths, kernel sizes, strides, or whether convolutions run on 2-D raster maps versus 1×1 (pointwise) features are given. Without these details the central claim that spatial correlation is what drives the Chamfer gains cannot be independently verified or reimplemented.
  2. [§3.3, Tables 1–3] §3.3 defines five loss-balancing constants (λ_img, λ_normal, λ_mask, λ_SSIM, λ_LPIPS) and an Adam schedule with mesh-LR decay on subdivision, yet no numerical values or selection procedure appear anywhere in the manuscript or tables. These free parameters directly control the trade-off between geometry regularizers and image terms that produce the reported meshes; their omission is a load-bearing reproducibility gap for the quantitative claims in Tables 1–3.
  3. [§3, §5, Fig. 8] The pipeline’s dependence on a coarse initial mesh inside a mask-defined visual hull (§3 opening, SfSil, progressive subdivision) is acknowledged in §5 but never quantified. A sensitivity study (sphere vs. SfSil vs. noisy/incomplete masks; different subdivision schedules) is needed to show that the convolutional shader’s gradients recover true high-frequency geometry rather than view-consistent appearance that is only later baked into G. This is the weakest load-bearing assumption behind the dark-region and boundary claims.
minor comments (5)
  1. [§4.2.3, Fig. 1, Fig. 6] Typographical errors: “T exturelss” (heading 4.2.3), “B oundary” (Fig. 1 caption), “Neuralanglo” (multiple places), “Chamf erdistances” (Fig. 6 discussion).
  2. [Table 1] Table 1 caption and body use “×10^{-3}” inconsistently with the prose; clarify units once and keep them uniform.
  3. [Fig. 2] Fig. 2 pipeline diagram is helpful but does not indicate where positional encoding, diffuse/specular split, or loss terms attach; a slightly more annotated version would aid readers.
  4. No mention of code or trained-model release. For a methods paper whose gains rest on architectural and hyperparameter choices, a public implementation would substantially strengthen the contribution.
  5. [§2.2] Related-work coverage of recent mesh/Gaussian hybrids is thin beyond SuGaR; a short paragraph situating CNS against other deferred-shading or rasterization-based neural meshes would help.

Circularity Check

0 steps flagged

No circularity: empirical multi-view reconstruction method evaluated against external GT benchmarks

full rationale

The paper proposes an architectural pipeline (mesh representation + fine-detail displacement network f_ϕ + convolutional neural shader f_θ, Eqs. 1–3) trained by joint optimization under standard image/mask/Laplacian/normal/SSIM/LPIPS losses against multi-view photographs. Reported gains (DTU mean Chamfer 0.49 vs. Neuralangelo 0.61; THuman dark subset 0.41 vs. 0.72; ablations in Table 3) are direct numerical comparisons of the optimized mesh G to public ground-truth point clouds and of renders to held-out images. Nothing is defined in terms of the quantity being predicted, no parameter fitted on a subset is re-labeled a prediction, and no load-bearing uniqueness or ansatz is imported via self-citation. The initial-mesh/visual-hull prior is an explicit modeling assumption (§3, §5), not a circular reduction. The derivation chain is ordinary supervised reconstruction and is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 2 invented entities

Load-bearing content is architectural and empirical, not axiomatic physics. The claim rests on standard differentiable-rendering assumptions, public multi-view datasets with masks/cameras, hand-chosen loss weights and network hyperparameters, and two invented modules (conv shader, displacement net) whose value is shown only via ablations inside this paper.

free parameters (4)
  • λ_img, λ_normal, λ_mask, λ_SSIM, λ_LPIPS (loss balancers)
    Eqs. 4–8 introduce balancing constants that control mesh vs image terms and perceptual terms; numeric values are never reported, so they are free knobs fitted for reported tables.
  • Adam learning rate and mesh LR decay schedule = 1e-4, half on upsample
    §4.1 sets LR=1e-4 and halves mesh LR on each subdivision; schedule is a hand-chosen training hyperparameter the geometry quality depends on.
  • positional encoding levels = 4
    §4.1 fixes Fourier feature levels at 4 for mesh positions; changes expressivity of the shader.
  • mesh subdivision frequency / target resolution
    §3 states mesh is upsampled at regular intervals; cadence and final vertex count are free design choices affecting Chamfer fairness vs implicit methods.
axioms (4)
  • domain assumption Differentiable rasterization of a triangle mesh plus barycentric interpolation yields usable position/normal maps for gradient-based inverse rendering.
    Pipeline premise in §3 and Fig. 2; inherited from NDS-style mesh deferred shading.
  • domain assumption Object masks and calibrated cameras are available so the mesh can be constrained to the visual hull (mask loss Eq. 10) and an SfSil or primitive init is adequate.
    §3 input requirements and §4.1 datasets; fails when masks/occlusion are poor (§5).
  • ad hoc to paper Local convolutional context over (γ(p_fine), n_fine, v) improves geometry disambiguation versus pointwise MLPs in dark/textureless regions.
    Central modeling bet of §3.2; supported only by this paper’s ablations/comparisons, not a prior theorem.
  • domain assumption Laplacian and normal-consistency mesh regularizers (Eqs. 6–7) do not erase the high-frequency detail the displacement network is meant to recover.
    Standard mesh-opt prior in §3.3.1; tension with fine-detail claim is unresolved analytically.
invented entities (2)
  • Convolutional neural shader (CNS shader) no independent evidence
    purpose: Predict pixel color from locally correlated fine positions/normals/view direction instead of single-point MLP shading.
    Named module in §3.2; novelty relative to NDS/NeRF MLPs is the conv inductive bias. No existence claim beyond the trained network.
  • Fine-detail displacement network no independent evidence
    purpose: Predict Δp and Δn in rendering coordinates so coarse rasterized meshes carry high-frequency geometry.
    Eqs. 1–2 in §3.1; architectural invention evaluated only via internal ablation.

pith-pipeline@v1.2.0-daily-grok45 · 19105 in / 3413 out tokens · 70700 ms · 2026-07-31T16:50:56.852492+00:00 · methodology

0 comments
read the original abstract

We propose a convolutional neural shading (CNS), a novel pipeline to reconstruct high-quality 3D shapes from multi-view images. Several recent studies have used neural radiance fields and other neural differentiable rendering methods to understand 3D geometry. However, these approaches rely on single-point geometric information, such as positions and normals of the surface, leading to a lack of detailed local geometry. Our approach addresses the inherent limitations of single-point information by leveraging a neural shader to capture variations even in dark and textureless regions with a convolutional neural shader, resulting in far more accurate geometry predictions. Additionally, our method mitigates surface irregularities at image boundaries by introducing a fine-detail displacement network, which utilizes spatial information of surface geometry and learns fine displacement details by correlating neighboring values in the rendering coordinates. Through extensive experiments, our proposed method has demonstrated significant quality improvements in the reconstructed shapes and rendered images over current state-of-the-art methods.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 3 linked inside Pith

  1. [1]

    Multimedia Systems 30(1), 24 (2024)

    Sarmah, M., Neelima, A.: Enhanced 3d reconstruction with all-neighbor-first philosophy and ricci flow-based mesh smoothing approach. Multimedia Systems 30(1), 24 (2024)

    2024

  2. [2]

    Multimedia Systems 28(5), 1713–1722 (2022)

    Zhu, L., Wang, S., Zhao, Z., Xu, X., Liu, Q.: Ced-net: contextual encoder–decoder network for 3d face reconstruction. Multimedia Systems 28(5), 1713–1722 (2022)

    2022

  3. [3]

    arXiv preprint arXiv:2210.00379 (2022)

    Gao, K., Gao, Y., He, H., Lu, D., Xu, L., Li, J.: NeRF: Neural radiance field in 3d vision, a comprehensive review. arXiv preprint arXiv:2210.00379 (2022)

    Pith/arXiv arXiv 2022

  4. [4]

    Multimedia Systems 30(4), 235 (2024)

    Huang, D., Shi, Y., Liu, J., Tang, W.: Self-supervised learning for fine-grained monocular 3d face reconstruction in the wild. Multimedia Systems 30(4), 235 (2024)

    2024

  5. [5]

    Multimedia Systems 29(3), 1579–1592 (2023)

    Liu, L., Gao, Y., Sun, J., Chen, J.: Single-image clothed 3d human reconstruction guided by a well-aligned parametric body model. Multimedia Systems 29(3), 1579–1592 (2023)

    2023

  6. [6]

    IEEE Transactions on Image Processing 22(5), 1901–1914 (2013) 20

    Shen, S.: Accurate multiple view 3d reconstruction using patch-based stereo for large-scale scenes. IEEE Transactions on Image Processing 22(5), 1901–1914 (2013) 20

    1901

  7. [7]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Delaunoy, A., Pollefeys, M.: Photometric bundle adjustment for dense multi-view 3d modeling. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1486–1493 (2014)

    2014

  8. [8]

    Journal of King Saud University- Computer and Information Sciences 34(5), 1663–1673 (2022)

    Hamid, M.S., Abd Manap, N., Hamzah, R.A., Kadmin, A.F.: Stereo matching algorithm based on deep learning: A survey. Journal of King Saud University- Computer and Information Sciences 34(5), 1663–1673 (2022)

    2022

  9. [9]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 44(4), 1738–1764 (2020)

    Laga, H., Jospin, L.V., Boussaid, F., Bennamoun, M.: A survey on deep learn- ing techniques for stereo-based depth estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence 44(4), 1738–1764 (2020)

    2020

  10. [10]

    International Archives of Photogrammetry and Remote Sensing 27(B8), 11–23 (1988)

    Lemmens, M.: A survey on stereo matching techniques. International Archives of Photogrammetry and Remote Sensing 27(B8), 11–23 (1988)

    1988

  11. [11]

    In: Proceedings of the 15th ACM Multimedia Systems Conference, pp

    Skartados, E., Yucel, M.K., Manganelli, B., Drosou, A., Sa` a-Garriga, A.: Finding waldo: Towards efficient exploration of nerf scene spaces. In: Proceedings of the 15th ACM Multimedia Systems Conference, pp. 155–165 (2024)

    2024

  12. [12]

    Multimedia Systems 30(5), 266 (2024)

    Dong, W., Liu, J., Chen, L., Sun, W., Pan, X., Ke, Y.: Implicit neural rep- resentation steganography by neuron pruning. Multimedia Systems 30(5), 266 (2024)

    2024

  13. [13]

    In: Proceedings of the Advances in Neural Information Processing Systems, vol

    Wang, Y., Skorokhodov, I., Wonka, P.: HF-NeuS: Improved surface reconstruc- tion using high-frequency details. In: Proceedings of the Advances in Neural Information Processing Systems, vol. 35, pp. 1966–1978 (2022)

    1966

  14. [14]

    Communications of the ACM 65(1), 99–106 (2021)

    Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: NeRF: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM 65(1), 99–106 (2021)

    2021

  15. [15]

    In: Proceedings of the Advances in Neural Information Processing Systems, vol

    Yariv, L., Kasten, Y., Moran, D., Galun, M., Atzmon, M., Ronen, B., Lipman, Y.: Multiview neural surface reconstruction by disentangling geometry and appear- ance. In: Proceedings of the Advances in Neural Information Processing Systems, vol. 33, pp. 2492–2502 (2020)

    2020

  16. [16]

    In: Proceedings of the Advances in Neural Information Processing Systems, vol

    Wang, P., Liu, L., Liu, Y., Theobalt, C., Komura, T., Wang, W.: NeuS: Learning neural implicit surfaces by volume rendering for multi-view Reconstruction. In: Proceedings of the Advances in Neural Information Processing Systems, vol. 34, pp. 27171–27183 (2021)

    2021

  17. [17]

    In: Proceedings of the Advances in Neural Information Processing Systems, vol

    Yariv, L., Gu, J., Kasten, Y., Lipman, Y.: Volume rendering of neural implicit sur- faces. In: Proceedings of the Advances in Neural Information Processing Systems, vol. 34, pp. 4805–4815 (2021)

    2021

  18. [18]

    In: Proceedings of 21 the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Worchel, M., Diaz, R., Hu, W., Schreer, O., Feldmann, I., Eisert, P.: Multi- view mesh reconstruction with neural deferred shading. In: Proceedings of 21 the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6187–6197 (2022)

    2022

  19. [19]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Long, X., Lin, C., Liu, L., Liu, Y., Wang, P., Theobalt, C., Komura, T., Wang, W.: Neuraludf: Learning unsigned distance fields for multi-view reconstruction of surfaces with arbitrary topologies. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 20834–20843 (2023)

    2023

  20. [20]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Li, Z., M¨ uller, T., Evans, A., Taylor, R.H., Unberath, M., Liu, M.-Y., Lin, C.- H.: Neuralangelo: High-fidelity neural surface reconstruction. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8456–8465 (2023)

    2023

  21. [21]

    Acta Numerica 26, 305–364 (2017)

    ¨Ozye¸ sil, O., Voroninski, V., Basri, R., Singer, A.: A survey of structure from motion. Acta Numerica 26, 305–364 (2017)

    2017

  22. [22]

    Artificial Intelligence Review 43, 55–81 (2015)

    Fuentes-Pacheco, J., Ruiz-Ascencio, J., Rend´ on-Mancha, J.M.: Visual simultane- ous localization and mapping: A survey. Artificial Intelligence Review 43, 55–81 (2015)

    2015

  23. [23]

    Displays 70, 102102 (2021)

    Wang, X., Wang, C., Liu, B., Zhou, X., Zhang, L., Zheng, J., Bai, X.: Multi-view stereo in the deep learning era: A comprehensive review. Displays 70, 102102 (2021)

    2021

  24. [24]

    arXiv preprint arXiv:2106.15328 (2021)

    Zhu, Q., Min, C., Wei, Z., Chen, Y., Wang, G.: Deep learning for multi-view stereo via plane sweep: A survey. arXiv preprint arXiv:2106.15328 (2021)

    Pith/arXiv arXiv 2021

  25. [25]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Liu, J., Ji, P., Bansal, N., Cai, C., Yan, Q., Huang, X., Xu, Y.: PlaneMVS: 3D plane reconstruction from multi-view stereo. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8665–8675 (2022)

    2022

  26. [26]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Wang, D., Cui, X., Chen, X., Zou, Z., Shi, T., Salcudean, S., Wang, Z.J., Ward, R.: Multi-view 3d reconstruction with transformers. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 5722–5731 (2021)

    2021

  27. [27]

    Computational Intelligence and Neuroscience 2020 (2020)

    Zhou, K., Meng, X., Cheng, B.: Review of stereo matching algorithms based on deep learning. Computational Intelligence and Neuroscience 2020 (2020)

    2020

  28. [28]

    In: International Conference on Learning Representations (2021)

    Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., Houlsby, N.: An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learning Representations (2021)

    2021

  29. [29]

    International Journal of Computer Vision 62, 221–247 (2005) 22

    Cheung, K.-M., Baker, S., Kanade, T.: Shape-from-silhouette across time part i: Theory and algorithms. International Journal of Computer Vision 62, 221–247 (2005) 22

    2005

  30. [30]

    In: Proceedings of the Advances in Neural Information Processing Systems, vol

    Tancik, M., Srinivasan, P., Mildenhall, B., Fridovich-Keil, S., Raghavan, N., Sing- hal, U., Ramamoorthi, R., Barron, J., Ng, R.: Fourier features let networks learn high frequency functions in low dimensional domains. In: Proceedings of the Advances in Neural Information Processing Systems, vol. 33, pp. 7537–7547 (2020)

    2020

  31. [31]

    IEEE Transactions on Image Processing 13(4), 600–612 (2004)

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assess- ment: From error visibility to structural similarity. IEEE Transactions on Image Processing 13(4), 600–612 (2004)

    2004

  32. [32]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 586–595 (2018)

    2018

  33. [33]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    Pith/arXiv arXiv 2014

  34. [34]

    In: Proceedings of the Eurographics/ACM SIGGRAPH Symposium on Geometry Processing, pp

    Botsch, M., Kobbelt, L.: A remeshing approach to multiresolution modeling. In: Proceedings of the Eurographics/ACM SIGGRAPH Symposium on Geometry Processing, pp. 185–192 (2004)

    2004

  35. [35]

    In: Proceedings of the IEEE International Conference on Computer Vision, pp

    He, K., Zhang, X., Ren, S., Sun, J.: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1026–1034 (2015)

    2015

  36. [36]

    In: Proceedings of the Advances in Neural Information Processing Systems Workshop (2017)

    Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., Lerer, A.: Automatic differentiation in PyTorch. In: Proceedings of the Advances in Neural Information Processing Systems Workshop (2017)

    2017

  37. [37]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Jensen, R., Dahl, A., Vogiatzis, G., Tola, E., Aanæs, H.: Large scale multi- view stereopsis evaluation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 406–413 (2014)

    2014

  38. [38]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Yao, Y., Luo, Z., Li, S., Zhang, J., Ren, Y., Zhou, L., Fang, T., Quan, L.: BlendedMVS: A large-scale dataset for generalized multi-view stereo networks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1790–1799 (2020)

    2020

  39. [39]

    In: Proceedings of the European Conference on Computer Vision, pp

    Shao, R., Zheng, Z., Zhang, H., Sun, J., Liu, Y.: DiffuStereo: High quality human reconstruction via diffusion-based stereo using sparse cameras. In: Proceedings of the European Conference on Computer Vision, pp. 702–720 (2022). Springer

    2022

  40. [40]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Gu´ edon, A., Lepetit, V.: SuGaR: Surface-aligned gaussian splatting for efficient 3D mesh reconstruction and high-quality mesh rendering. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5354– 5363 (2024) 23

    2024

  41. [41]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016)

    Schonberger, J.L., Frahm, J.-M.: Structure-from-motion revisited. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2016)

    2016

  42. [42]

    ACM Transactions on Graphics (TOG) 42(4), 1–14 (2023)

    Kerbl, B., Kopanas, G., Leimkuehler, T., Drettakis, G.: 3D Gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics (TOG) 42(4), 1–14 (2023)

    2023

  43. [43]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Zheng, Y., Shao, R., Zhang, Y., Yu, T., Zheng, Z., Dai, Q., Liu, Y.: DeepMultiCap: Performance capture of multiple characters using sparse multiview cameras. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6239–6249 (2021) 24

    2021