pith. sign in

arxiv: 2606.30677 · v1 · pith:7M65PUSXnew · submitted 2026-06-26 · 💻 cs.GR · cs.CV

DANTE-W: Diffuse Albedo Neural Texturing in the Wild

Guangyu Wang , Tianheng Lu , Ruqi Huang , Lu Fang This is my paper

Pith reviewed 2026-07-01 06:59 UTC · model grok-4.3

classification 💻 cs.GR cs.CV
keywords diffuse albedoneural texturingmesh parameterizationin-the-wild scenesneural renderingrelightingtexture recovery3D reconstruction
0
0 comments X

The pith

DANTE-W recovers high-fidelity diffuse albedo textures from unstructured image collections of large-scale in-the-wild scenes.

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

Classical mesh texturing blends multi-view images but bakes in shading and shadows that ruin relighting. DANTE-W introduces a neural framework that takes a reconstructed mesh plus its parameterization and fuses view-space generative albedo priors into one coherent texture map. Physically principled neural rendering then sharpens fine details while the output stays compatible with standard 3D pipelines. The authors test the approach on a new benchmark mixing real outdoor scenes and synthetic objects, showing cleaner albedo and improved relighting.

Core claim

Given a reconstructed mesh and its surface parameterization, DANTE-W fuses view-space generative albedo priors into a coherent texture space via an expressive neural representation while substantially enhancing fine-grained textural details through physically principled neural rendering, enabling high-fidelity diffuse albedo texture recovery from unstructured image collections for large-scale, in-the-wild scenes.

What carries the argument

Expressive neural representation that fuses view-space generative albedo priors into coherent texture space, paired with physically principled neural rendering for detail enhancement.

If this is right

  • Recovered textures contain no baked-in shading or cast shadows from the original captures.
  • The framework integrates directly with existing 3D reconstruction pipelines without requiring controlled lighting.
  • Fine textural details are preserved and enhanced while albedo remains physically consistent.
  • A new benchmark dataset of diverse real-world and synthetic scenes enables systematic evaluation of albedo recovery.
  • Relighting results improve in visual fidelity compared with direct image blending.

Where Pith is reading between the lines

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

  • The same fusion step could be applied to photogrammetry outputs for urban or heritage modeling where capture is entirely uncontrolled.
  • Extending the neural representation to additional material channels such as roughness might allow joint recovery of multiple surface properties.
  • If the mesh is updated over time, the method could support texture tracking in dynamic scenes without re-capturing full image sets.

Load-bearing premise

The input mesh and its surface parameterization are accurate and complete enough that view priors can be fused without introducing topological or parameterization artifacts.

What would settle it

Render the recovered texture under new lighting conditions and check whether residual shadows or view-dependent lighting from the input images remain visible; clean separation would confirm the claim.

Figures

Figures reproduced from arXiv: 2606.30677 by Guangyu Wang, Lu Fang, Ruqi Huang, Tianheng Lu.

Figure 1
Figure 1. Figure 1: We present Dante-w, a neural texturing framework for high-fidelity diffuse albedo recovery in the wild. Compared to vanilla mesh texturing with baked-in lighting effects (e.g., noon-time shading and strong roof-edge shadowing on this pavilion), our method effectively disentangles a 3D-consistent diffuse albedo texture with exceptional photorealism. Leveraging physically principled neural rendering, Dante-w… view at source ↗
Figure 2
Figure 2. Figure 2: An overview of our physically principled neural rendering framework. Given a reconstructed mesh of the scene with surface parameterization, we represent diffuse albedo texture ad using a high-resolution 2D hash encoding ψ 2D a (·) and irradiance sd using a low-resolution 3D hash encoding ψ 3D s (·). This explicit frequency-band discrep￾ancy (with the maximal grid resolutions satisfying V (a) La ≫ V (s) Ls … view at source ↗
Figure 3
Figure 3. Figure 3: An illustration of the pro￾posed hash-encoded neural texture representation. Specifically, we apply multi-resolution hash encoding [49] in the 2D texture space with a hierarchy of spatial grid resolutions {V (a) ℓ } La ℓ=1. We then use a lightweight MLP, FΘa , to interpret the hash features and de￾code the diffuse albedo value. Let xˆ ∈ R 2 be the parametric UV-coordinate of a surface position, we denote b… view at source ↗
Figure 4
Figure 4. Figure 4: Visual comparison of diffuse albedo recovery on in-the-wild scenes. By lifting view-space diffusion priors upon a coherent neural texture via physically principled neural rendering (PR), our method robustly eliminates baked-in shading and shadowing effects that remain challenging for vanilla mesh texturing (Metashape Pro [45]), while refining raw diffusion outputs of intrinsic decomposition models (RGB↔X [… view at source ↗
Figure 5
Figure 5. Figure 5: Generalization on sharp shadows and strong specularity using GigaLit scenes. texture. For recent diffusion-based renderers such as Cosmos-DiffusionRenderer (DR) [38] and RGB↔X [66], their raw diffusion outputs primarily lack accuracy, stability and 3D consistency across varying viewpoints, particularly at fine de￾tails. By contrast, our method enables strictly 3D-consistent and more faithful diffuse albedo… view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative results of diffuse albedo recovery and relighting on GigaLit syn￾thetic objects, compared with vanilla texturing (Metashape Pro [45]), raw outputs of RGB↔X [66] and Cosmos-DiffusionRenderer (DR) [38], and the ablated variant of our method in terms of physically principled neural rendering (PR). The albedo is recovered under the Original Lit and subsequently relighted under the Novel Lit. Our me… view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative comparisons on diffuse albedo recovery using Stanford-ORB dataset [34]. The raw diffusion outputs from both DiffusionRenderer (DR) [38] and RGB↔X [66] tend to exhibit hallucinated artifacts caused by the probabilistic nature of generative modelling and the information loss of the VAE latent. Naively lifting view￾space priors without our physically principled neural rendering (Ours+DR/RGB↔X (w/o… view at source ↗
Figure 8
Figure 8. Figure 8: Visualizations of diffuse albedo recovery and relighting on a large-scale, in-the￾wild scene – The Pavilion of Prince Teng – reconstructed using over 2,500 unstructured photographs. Our method learns a consistent diffuse albedo texture with significantly finer details compared to the naive lifting of diffusion priors (w/o PR). We also demon￾strate superior fidelity compared to the vanilla texturing approac… view at source ↗
Figure 9
Figure 9. Figure 9: Ablation study on frequency discrepancy. Differentiating the maximal grid resolution facilitates the disentanglement [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Run time comparison with Metashape Pro [45] in terms of mesh texturing. Our method demonstrates comparable visual quality while achieving over 7× acceleration. for real-world scenes, the ground-truth albedo in Stanford-ORB serves only as a coarse reference, failing to provide a principled and rigorous quantitative evalu￾ation for the task of high-fidelity diffuse albedo recovery, particularly for highly i… view at source ↗
read the original abstract

Classical mesh texturing techniques blend captured multi-view images directly, which inevitably suffer from baked-in shading and casted shadows that compromise visual fidelity during relighting. To circumvent this issue, we present a neural texturing framework, namely DANTE-W, to enable high-fidelity diffuse albedo texture recovery from unstructured image collections for large-scale, in-the-wild scenes, which integrates seamlessly with traditional 3D reconstruction pipelines. Given a reconstructed mesh and its surface parameterization, our method fuses view-space generative albedo priors into a coherent texture space via an expressive neural representation, while substantially enhancing fine-grained textural details through physically principled neural rendering. To comprehensively evaluate our method, we curate a benchmark dataset featuring diverse, fine-grained textures, comprising both real-world in-the-wild scenes and synthetic objects. Extensive experiments verify the effectiveness of our approach in reconstructing accurate albedo textures and boosting relighting fidelity. Project page: dante-wild.github.io.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper presents DANTE-W, a neural texturing method that recovers high-fidelity diffuse albedo textures from unstructured multi-view image collections of large-scale in-the-wild scenes. Given a reconstructed mesh and its UV parameterization, the approach fuses view-space generative albedo priors into a coherent texture space via an expressive neural representation and applies physically principled neural rendering to enhance fine-grained details. It claims seamless integration with traditional 3D reconstruction pipelines and validates this on a new benchmark dataset of real-world and synthetic scenes with diverse textures, showing improved albedo accuracy and relighting fidelity.

Significance. If the central claims hold under the stated assumptions, the work would address a persistent limitation in classical mesh texturing by enabling shading-free albedo recovery at scale, potentially improving downstream applications such as relighting and material editing in unstructured captures. The curation of a dedicated benchmark with both real and synthetic data is a concrete contribution that could support future comparisons.

major comments (2)
  1. [Abstract] Abstract and method overview: The central claim of high-fidelity albedo recovery that 'integrates seamlessly with traditional 3D reconstruction pipelines' is load-bearing on the premise that the input mesh and surface parameterization are sufficiently accurate and complete. No experiments, ablations, or robustness tests are described that evaluate performance under typical SfM/MVS defects (holes, topological errors, or parameterization distortions) common in in-the-wild scenes; this leaves the fusion step's behavior on imperfect inputs unverified.
  2. [Evaluation] Evaluation section (implied by benchmark description): While a new benchmark is introduced, the abstract provides no quantitative metrics, ablation studies, or comparisons that isolate the contribution of the neural fusion versus view selection or post-processing; without these, it is not possible to confirm that the reported improvements in albedo accuracy stem from the proposed representation rather than dataset curation or hyperparameter choices.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. Below we respond point-by-point to the major comments, clarifying the scope of our claims while committing to targeted revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract] Abstract and method overview: The central claim of high-fidelity albedo recovery that 'integrates seamlessly with traditional 3D reconstruction pipelines' is load-bearing on the premise that the input mesh and surface parameterization are sufficiently accurate and complete. No experiments, ablations, or robustness tests are described that evaluate performance under typical SfM/MVS defects (holes, topological errors, or parameterization distortions) common in in-the-wild scenes; this leaves the fusion step's behavior on imperfect inputs unverified.

    Authors: The phrase 'integrates seamlessly' denotes that DANTE-W is a modular post-process accepting any mesh and UV map produced by an upstream SfM/MVS pipeline, without requiring changes to the reconstruction code or camera calibration. The experiments are performed on real-world captures whose meshes already contain the typical defects of in-the-wild reconstruction; however, we did not isolate performance under controlled defect injection. We will add a dedicated robustness subsection with qualitative and quantitative analysis on meshes with simulated holes and parameterization distortions, together with a short discussion of failure modes. revision: yes

  2. Referee: [Evaluation] Evaluation section (implied by benchmark description): While a new benchmark is introduced, the abstract provides no quantitative metrics, ablation studies, or comparisons that isolate the contribution of the neural fusion versus view selection or post-processing; without these, it is not possible to confirm that the reported improvements in albedo accuracy stem from the proposed representation rather than dataset curation or hyperparameter choices.

    Authors: Abstracts conventionally omit numerical tables. The evaluation section of the manuscript already contains (i) quantitative albedo and relighting metrics against multiple baselines, (ii) component ablations that disable the generative prior, the neural renderer, or both, and (iii) controlled comparisons that keep the benchmark fixed while varying only the fusion representation. These results are designed to attribute gains to the neural texture representation rather than dataset choice. If the referee finds the isolation still insufficient, we will expand the ablation table with an additional row that replaces our fusion module by simple view-selection averaging while keeping all other factors identical. revision: partial

Circularity Check

0 steps flagged

No circularity in derivation; method is input-driven without self-referential reductions

full rationale

The provided abstract and description outline a neural texturing pipeline that takes a reconstructed mesh and surface parameterization as given inputs, then fuses generative priors via a neural representation and applies neural rendering for detail enhancement. No equations, predictions, or first-principles derivations are shown that reduce outputs to fitted inputs by construction, nor any self-citation chains, uniqueness theorems, or ansatzes that collapse the central claim. The approach is presented as a novel integration with existing 3D pipelines rather than a tautological renaming or self-definition, making the derivation self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only; no equations or implementation details available to enumerate free parameters, axioms, or invented entities. The method implicitly assumes the existence of reliable generative albedo priors and a physically principled renderer, but these cannot be audited without the full text.

pith-pipeline@v0.9.1-grok · 5693 in / 1159 out tokens · 24984 ms · 2026-07-01T06:59:18.157554+00:00 · methodology

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

70 extracted references · 70 canonical work pages · 2 internal anchors

  1. [1]

    Cosmos World Foundation Model Platform for Physical AI

    Agarwal, N., Ali, A., Bala, M., Balaji, Y., Barker, E., Cai, T., Chattopadhyay, P., Chen, Y., Cui, Y., Ding, Y., et al.: Cosmos world foundation model platform for physical ai. arXiv preprint arXiv:2501.03575 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  2. [2]

    Cosmos-transfer1: Conditional world generation with adaptive multimodal control, 2025.https://arxiv.org/abs/2503.14492

    Alhaija, H.A., Alvarez, J., Bala, M., Cai, T., Cao, T., Cha, L., Chen, J., Chen, M., Ferroni, F., Fidler, S., et al.: Cosmos-transfer1: Conditional world generation with adaptive multimodal control. arXiv preprint arXiv:2503.14492 (2025)

    work page arXiv 2025

  3. [3]

    In: Proceedings of the Computer Vision and Pattern Recognition Con- ference

    Alzayer, H., Henzler, P., Barron, J.T., Huang, J.B., Srinivasan, P.P., Verbin, D.: Generative multiview relighting for 3d reconstruction under extreme illumination variation. In: Proceedings of the Computer Vision and Pattern Recognition Con- ference. pp. 10933–10942 (2025)

    work page 2025

  4. [4]

    In: Color and imaging conference

    Anderson, M., Motta, R., Chandrasekar, S., Stokes, M.: Proposal for a standard default color space for the internet—srgb. In: Color and imaging conference. vol. 4, pp. 238–245. Society of Imaging Science and Technology (1996)

    work page 1996

  5. [5]

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

    Bai,H.,Zhu,J.,Jiang,S.,Huang,W.,Lu,T.,Li,Y.,Guo,J.,Fu,R.,Guo,Y.,Chen, L.: Gare: Relightable 3d gaussian splatting for outdoor scenes from unconstrained photo collections. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 26456–26465 (2025)

    work page 2025

  6. [6]

    ICCV (2023)

    Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Zip-nerf: Anti-aliased grid-based neural radiance fields. ICCV (2023)

    work page 2023

  7. [7]

    ACM Transactions on Graphics (TOG)33(4), 1–12 (2014)

    Bell, S., Bala, K., Snavely, N.: Intrinsic images in the wild. ACM Transactions on Graphics (TOG)33(4), 1–12 (2014)

    work page 2014

  8. [8]

    ACM Trans

    Bi, S., Kalantari, N.K., Ramamoorthi, R.: Patch-based optimization for image- based texture mapping. ACM Trans. Graph.36(4), 106–1 (2017)

    work page 2017

  9. [9]

    Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets

    Blattmann, A., Dockhorn, T., Kulal, S., Mendelevitch, D., Kilian, M., Lorenz, D., Levi, Y., English, Z., Voleti, V., Letts, A., et al.: Stable video diffusion: Scaling latent video diffusion models to large datasets. arXiv preprint arXiv:2311.15127 (2023) 16 G. Wang et al

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  10. [10]

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

    Boss, M., Braun, R., Jampani, V., Barron, J.T., Liu, C., Lensch, H.: Nerd: Neural reflectance decomposition from image collections. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12684–12694 (2021)

    work page 2021

  11. [11]

    Advances in Neural Information Processing Systems34, 10691–10704 (2021)

    Boss, M., Jampani, V., Braun, R., Liu, C., Barron, J., Lensch, H.: Neural-pil: Neural pre-integrated lighting for reflectance decomposition. Advances in Neural Information Processing Systems34, 10691–10704 (2021)

    work page 2021

  12. [12]

    In: Acm siggraph

    Burley, B., Studios, W.D.A.: Physically-based shading at disney. In: Acm siggraph. vol. 2012, pp. 1–7. vol. 2012 (2012)

    work page 2012

  13. [13]

    ACM Transactions on Graphics (TOG)43(6), 1–12 (2024)

    Careaga, C., Aksoy, Y.: Colorful diffuse intrinsic image decomposition in the wild. ACM Transactions on Graphics (TOG)43(6), 1–12 (2024)

    work page 2024

  14. [14]

    ACM Trans

    Careaga, C., Aksoy, Y.: Intrinsic image decomposition via ordinal shading. ACM Trans. Graph.43(1) (2023)

    work page 2023

  15. [15]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Chang, Y., Kim, Y., Seo, S., Yi, J., Kwak, N.: Fast sun-aligned outdoor scene relighting based on tensorf. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 3626–3636 (2024)

    work page 2024

  16. [16]

    IEEE Transactions on Visualization and Computer Graphics (2024)

    Chen, D., Li, H., Ye, W., Wang, Y., Xie, W., Zhai, S., Wang, N., Liu, H., Bao, H., Zhang, G.: Pgsr: Planar-based gaussian splatting for efficient and high-fidelity sur- face reconstruction. IEEE Transactions on Visualization and Computer Graphics (2024)

    work page 2024

  17. [17]

    In: European Conference on Computer Vision

    Chen, X., Peng, S., Yang, D., Liu, Y., Pan, B., Lv, C., Zhou, X.: Intrinsicanything: Learning diffusion priors for inverse rendering under unknown illumination. In: European Conference on Computer Vision. pp. 450–467. Springer (2024)

    work page 2024

  18. [18]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Deitke, M., Schwenk, D., Salvador, J., Weihs, L., Michel, O., VanderBilt, E., Schmidt, L., Ehsani, K., Kembhavi, A., Farhadi, A.: Objaverse: A universe of annotated 3d objects. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13142–13153 (2023)

    work page 2023

  19. [19]

    In: Proceedings of the Thirty-Fourth Interna- tional Joint Conference on Artificial Intelligence

    Feng, W., Ye, K., Zhang, Q., Zhang, Q., Li, N.: 2d gaussian splatting for outdoor scene decomposition and relighting. In: Proceedings of the Thirty-Fourth Interna- tional Joint Conference on Artificial Intelligence. pp. 990–998 (2025)

    work page 2025

  20. [20]

    Foundation, T.B.: Blender 3.6 (2023)

    work page 2023

  21. [21]

    Foundations and trends®in Computer Graphics and Vision9(1-2), 1–148 (2015)

    Furukawa, Y., Hernández, C., et al.: Multi-view stereo: A tutorial. Foundations and trends®in Computer Graphics and Vision9(1-2), 1–148 (2015)

    work page 2015

  22. [22]

    In: European Conference on Computer Vision

    Gao, J., Gu, C., Lin, Y., Li, Z., Zhu, H., Cao, X., Zhang, L., Yao, Y.: Relightable 3d gaussians: Realistic point cloud relighting with brdf decomposition and ray tracing. In: European Conference on Computer Vision. pp. 73–89. Springer (2024)

    work page 2024

  23. [23]

    In: European conference on computer vision

    Gardner, J.A., Kashin, E., Egger, B., Smith, W.A.: The sky’s the limit: Relightable outdoor scenes via a sky-pixel constrained illumination prior and outside-in visi- bility. In: European conference on computer vision. pp. 126–143. Springer (2024)

    work page 2024

  24. [24]

    Greg, Z., Rob, T., Rico, C., James, R.C., Dario, B., Jenelle, v.H., et al.: Polyhaven: a curated public asset library for visual effects artists and game designers (2021)

    work page 2021

  25. [25]

    Advances in Neural Information Processing Systems35, 22856–22869 (2022)

    Hasselgren, J., Hofmann, N., Munkberg, J.: Shape, light, and material decomposi- tion from images using monte carlo rendering and denoising. Advances in Neural Information Processing Systems35, 22856–22869 (2022)

    work page 2022

  26. [26]

    Remote Sensing10(11), 1725 (2018)

    Huang, X., Zhu, Q., Jiang, W.: Gpvc: Graphics pipeline-based visibility classifica- tion for texture reconstruction. Remote Sensing10(11), 1725 (2018)

    work page 2018

  27. [27]

    Packt Publishing Ltd (2013)

    Iraci, B.: Blender cycles: lighting and rendering cookbook. Packt Publishing Ltd (2013)

    work page 2013

  28. [28]

    Advances in Neural Information Processing Systems37, 141129–141152 (2024) DANTE-W 17

    Jin, H., Li, Y., Luan, F., Xiangli, Y., Bi, S., Zhang, K., Xu, Z., Sun, J., Snavely, N.: Neural gaffer: Relighting any object via diffusion. Advances in Neural Information Processing Systems37, 141129–141152 (2024) DANTE-W 17

    work page 2024

  29. [29]

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

    Jin, H., Liu, I., Xu, P., Zhang, X., Han, S., Bi, S., Zhou, X., Xu, Z., Su, H.: Tensoir: Tensorial inverse rendering. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 165–174 (2023)

    work page 2023

  30. [30]

    Kajiya,J.T.:Therenderingequation.In:Proceedingsofthe13thannualconference on Computer graphics and interactive techniques. pp. 143–150 (1986)

    work page 1986

  31. [31]

    In: 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

    Kaleta, J., Kania, K., Trzciński, T., Kowalski, M.: Lumigauss: relightable gaussian splatting in the wild. In: 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). pp. 1–10. IEEE (2025)

    work page 2025

  32. [32]

    Karis, B., Games, E.: Real shading in unreal engine 4. Proc. Physically Based Shading Theory Practice4(3), 1 (2013)

    work page 2013

  33. [33]

    ACM Trans

    Kerbl, B., Kopanas, G., Leimkühler, T., Drettakis, G.: 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph.42(4), 139–1 (2023)

    work page 2023

  34. [34]

    In: Advances in Neural Information Processing Systems Datasets and Benchmarks Track (2023)

    Kuang, Z., Zhang, Y., Yu, H.X., Agarwala, S., Wu, E., Wu, J., et al.: Stanford- orb: a real-world 3d object inverse rendering benchmark. In: Advances in Neural Information Processing Systems Datasets and Benchmarks Track (2023)

    work page 2023

  35. [35]

    In: 2007 IEEE conference on computer vision and pattern recognition

    Lempitsky, V., Ivanov, D.: Seamless mosaicing of image-based texture maps. In: 2007 IEEE conference on computer vision and pattern recognition. pp. 1–6. IEEE (2007)

    work page 2007

  36. [36]

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

    Li, Z., Müller, 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)

    work page 2023

  37. [37]

    arXiv preprint arXiv:2412.12083 (2024)

    Li, Z., Wu, T., Tan, J., Zhang, M., Wang, J., Lin, D.: Idarb: Intrinsic decom- position for arbitrary number of input views and illuminations. arXiv preprint arXiv:2412.12083 (2024)

    work page arXiv 2024

  38. [38]

    In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2025)

    Liang, R., Gojcic, Z., Ling, H., Munkberg, J., Hasselgren, J., Lin, Z.H., Gao, J., Keller, A., Vijaykumar, N., Fidler, S., Wang, Z.: Diffusionrenderer: Neural inverse and forward rendering with video diffusion models. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2025)

    work page 2025

  39. [39]

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

    Liang, Z., Zhang, Q., Feng, Y., Shan, Y., Jia, K.: Gs-ir: 3d gaussian splatting for inverse rendering. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 21644–21653 (2024)

    work page 2024

  40. [40]

    arXiv preprint arXiv:2509.11275 (2025)

    Liao, L., Zhang, C., Wu, T., Lv, H., Deng, B., Gao, L.: Rosgs: Relightable outdoor scenes with gaussian splatting. arXiv preprint arXiv:2509.11275 (2025)

    work page arXiv 2025

  41. [41]

    IEEE transactions on geoscience and remote sensing61, 1–11 (2023)

    Ling, X., Qin, R.: Large-scale and efficient texture mapping algorithm via loopy belief propagation. IEEE transactions on geoscience and remote sensing61, 1–11 (2023)

    work page 2023

  42. [42]

    In: 2025 International Conference on 3D Vision (3DV)

    Litman, Y., Patashnik, O., Deng, K., Agrawal, A., Zawar, R., De la Torre, F., Tulsiani, S.: Materialfusion: Enhancing inverse rendering with material diffusion priors. In: 2025 International Conference on 3D Vision (3DV). pp. 802–812. IEEE (2025)

    work page 2025

  43. [43]

    In: Proceedings of the IEEE/CVF International Confer- ence on Computer Vision

    Litman, Y., De la Torre, F., Tulsiani, S.: Lightswitch: Multi-view relighting with material-guided diffusion. In: Proceedings of the IEEE/CVF International Confer- ence on Computer Vision. pp. 27750–27759 (2025)

    work page 2025

  44. [44]

    ACM Transactions on Graphics (ToG)42(4), 1–22 (2023)

    Liu, Y., Wang, P., Lin, C., Long, X., Wang, J., Liu, L., Komura, T., Wang, W.: Nero: Neural geometry and brdf reconstruction of reflective objects from multiview images. ACM Transactions on Graphics (ToG)42(4), 1–22 (2023)

    work page 2023

  45. [45]

    LLC, A.: Agisoft metashape professional (2023)

    work page 2023

  46. [46]

    Wang et al

    Martin-Brualla,R.,Radwan,N.,Sajjadi,M.S.,Barron,J.T.,Dosovitskiy,A.,Duck- worth, D.: Nerf in the wild: Neural radiance fields for unconstrained photo collec- 18 G. Wang et al. tions. In: Proceedings of theIEEE/CVF conference on computer vision and pattern recognition. pp. 7210–7219 (2021)

    work page 2021

  47. [47]

    Commu- nications of the ACM65(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. Commu- nications of the ACM65(1), 99–106 (2021)

    work page 2021

  48. [48]

    Müller, T.: tiny-cuda-nn (4 2021)

    work page 2021

  49. [49]

    ACM Transactions on Graphics (ToG)41(4), 1– 15 (2022)

    Müller,T.,Evans,A.,Schied,C.,Keller,A.:Instantneuralgraphicsprimitiveswith a multiresolution hash encoding. ACM Transactions on Graphics (ToG)41(4), 1– 15 (2022)

    work page 2022

  50. [50]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Munkberg, J., Hasselgren, J., Shen, T., Gao, J., Chen, W., Evans, A., Müller, T., Fidler, S.: Extracting Triangular 3D Models, Materials, and Lighting From Images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 8280–8290 (June 2022)

    work page 2022

  51. [51]

    Petr, D., Vilém, D., Karolína, H., Monika, R., Adam, K., et al.: Blenderkit: Ever- growing 3d asset library integrated into blender (2018)

    work page 2018

  52. [52]

    In: Computer Graphics Forum

    Poirier-Ginter, Y., Gauthier, A., Phillip, J., Lalonde, J.F., Drettakis, G.: A dif- fusion approach to radiance field relighting using multi-illumination synthesis. In: Computer Graphics Forum. vol. 43, p. e15147. Wiley Online Library (2024)

    work page 2024

  53. [53]

    In: Proceedings of the 28th annual conference on Computer graphics and interactive techniques

    Ramamoorthi, R., Hanrahan, P.: An efficient representation for irradiance environ- ment maps. In: Proceedings of the 28th annual conference on Computer graphics and interactive techniques. pp. 497–500 (2001)

    work page 2001

  54. [54]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Roberts, M., Ramapuram, J., Ranjan, A., Kumar, A., Bautista, M.A., Paczan, N., Webb, R., Susskind, J.M.: Hypersim: A photorealistic synthetic dataset for holis- tic indoor scene understanding. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 10912–10922 (2021)

    work page 2021

  55. [55]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B.: High-resolution image synthesis with latent diffusion models. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 10684–10695 (2022)

    work page 2022

  56. [56]

    In: European Conference on Computer Vision (ECCV) (2022)

    Rudnev, V., Elgharib, M., Smith, W., Liu, L., Golyanik, V., Theobalt, C.: Nerf for outdoor scene relighting. In: European Conference on Computer Vision (ECCV) (2022)

    work page 2022

  57. [57]

    In: Computer Vision–ECCV 2016: 14th Euro- pean Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part III 14

    Schönberger, J.L., Zheng, E., Frahm, J.M., Pollefeys, M.: Pixelwise view selection for unstructured multi-view stereo. In: Computer Vision–ECCV 2016: 14th Euro- pean Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part III 14. pp. 501–518. Springer (2016)

    work page 2016

  58. [58]

    In: Asian Conference on Computer Vision

    Shen, T., Wang, J., Fang, T., Zhu, S., Quan, L.: Color correction for image-based modeling in the large. In: Asian Conference on Computer Vision. pp. 392–407. Springer (2016)

    work page 2016

  59. [59]

    In: SIGGRAPH Asia 2023 Conference Papers (SA Conference Papers ’23) (2023)

    Sun, J.M., Wu, T., Yang, Y.L., Lai, Y.K., Gao, L.: Sol-nerf: Sunlight modeling for outdoor scene decomposition and relighting. In: SIGGRAPH Asia 2023 Conference Papers (SA Conference Papers ’23) (2023)

    work page 2023

  60. [60]

    In: The Thirty-ninth Annual Conference on Neural Information Processing Sys- tems (2025)

    Tang, J., Levine, M.J., Verbin, D., Garbin, S.J., Nießner, M., Brualla, R.M., Srini- vasan, P.P., Henzler, P.: Rogr: Relightable 3d objects using generative relighting. In: The Thirty-ninth Annual Conference on Neural Information Processing Sys- tems (2025)

    work page 2025

  61. [61]

    In: European conference on computer vision

    Waechter, M., Moehrle, N., Goesele, M.: Let there be color! large-scale texturing of 3d reconstructions. In: European conference on computer vision. pp. 836–850. Springer (2014)

    work page 2014

  62. [62]

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

    Wang, G., Zhang, J., Wang, F., Huang, R., Fang, L.: Xscale-nvs: Cross-scale novel view synthesis with hash featurized manifold. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 21029–21039 (2024) DANTE-W 19

    work page 2024

  63. [63]

    In: Proceedingsofthe35thInternationalConferenceonNeuralInformationProcessing Systems

    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: Proceedingsofthe35thInternationalConferenceonNeuralInformationProcessing Systems. pp. 27171–27183 (2021)

    work page 2021

  64. [64]

    arXiv preprint arXiv:2508.19789 (2025)

    Wu, X., Zhu, P., Lyu, J., Liu, X., Guo, J., Guo, Y., Xu, W., Lyu, C.: Stableintrin- sic: Detail-preserving one-step diffusion model for multi-view material estimation. arXiv preprint arXiv:2508.19789 (2025)

    work page arXiv 2025

  65. [65]

    In: European conference on computer vision

    Yao, Y., Zhang, J., Liu, J., Qu, Y., Fang, T., McKinnon, D., Tsin, Y., Quan, L.: Neilf: Neural incident light field for physically-based material estimation. In: European conference on computer vision. pp. 700–716. Springer (2022)

    work page 2022

  66. [66]

    ACM SIGGRAPH 2024 Conference Papers (2024), https://api.semanticscholar.org/CorpusID:269484178

    Zeng, Z., Deschaintre, V., Georgiev, I., Hold-Geoffroy, Y., Hu, Y., Luan, F., Yan, L.Q., Havs.an, M.: Rgbx: Image decomposition and synthesis using material- and lighting-aware diffusion models. ACM SIGGRAPH 2024 Conference Papers (2024), https://api.semanticscholar.org/CorpusID:269484178

    work page 2024

  67. [67]

    In: Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Zhang, K., Luan, F., Wang, Q., Bala, K., Snavely, N.: Physg: Inverse rendering with spherical gaussians for physics-based material editing and relighting. In: Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 5453–5462 (2021)

    work page 2021

  68. [68]

    ACM Transactions on Graphics (ToG)40(6), 1–18 (2021)

    Zhang, X., Srinivasan, P.P., Deng, B., Debevec, P., Freeman, W.T., Barron, J.T.: Nerfactor: Neural factorization of shape and reflectance under an unknown illumi- nation. ACM Transactions on Graphics (ToG)40(6), 1–18 (2021)

    work page 2021

  69. [69]

    Advances in Neural Information Processing Systems37, 42593–42617 (2024)

    Zhao, X., Srinivasan, P., Verbin, D., Park, K., Martin Brualla, R., Henzler, P.: Illuminerf: 3d relighting without inverse rendering. Advances in Neural Information Processing Systems37, 42593–42617 (2024)

    work page 2024

  70. [70]

    In: SIGGRAPH Asia 2022 Conference Papers

    Zhu, J., Luan, F., Huo, Y., Lin, Z., Zhong, Z., Xi, D., Wang, R., Bao, H., Zheng, J., Tang, R.: Learning-based inverse rendering of complex indoor scenes with dif- ferentiable monte carlo raytracing. In: SIGGRAPH Asia 2022 Conference Papers. pp. 1–8 (2022)

    work page 2022