pith. sign in

arxiv: 2606.29379 · v1 · pith:MOMNNKNUnew · submitted 2026-06-28 · 💻 cs.CV · cs.AI· cs.GR

DR-GS: Physically-Based Deformable and Relightable 2D Gaussians

Jiaxin Li , Tong Wu , Yi Wei , Tailin Wu , Li Zhang This is my paper

Pith reviewed 2026-06-30 07:02 UTC · model grok-4.3

classification 💻 cs.CV cs.AIcs.GR
keywords gaussian splattingdeformable objectsrelightingphysically-based renderinginverse renderingmaterial editing3D reconstructiondynamic scenes
0
0 comments X

The pith

DR-GS separates geometry, illumination, and material in 2D Gaussians to allow deformation and relighting without baked errors.

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

The paper presents DR-GS as a method that fixes two issues in existing Gaussian splatting for moving objects: illumination baked into textures that breaks under deformation or new lights, and the inability to edit materials after capture. It does this by building a single framework that performs physically-based inverse rendering while keeping the three elements separate. A reader would care because this separation produces appearance that stays consistent when conditions change and permits editing afterward. The result is claimed to work across static scenes, dynamic motion, and relighting tasks while keeping glossy reflections intact.

Core claim

DR-GS is a unified Gaussian framework that integrates physically-based inverse rendering, relighting, and deformation-aware manipulation through explicit disentanglement of geometry, illumination, and material representations, overcoming the limitations of static snapshots and enabling post-reconstruction parameter editing.

What carries the argument

Explicit disentanglement of geometry, illumination, and material representations inside the 2D Gaussian splatting pipeline

If this is right

  • Appearance stays physically consistent when objects deform or lighting changes.
  • Material parameters can be edited after the initial reconstruction is complete.
  • Reflections and specular highlights on glossy surfaces are preserved during both deformation and relighting.
  • A fully decoupled geometry-illumination-material pipeline supports high-quality 3D asset creation and editing.

Where Pith is reading between the lines

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

  • The same separation could be tested on other explicit scene representations to see if the consistency and editing benefits transfer.
  • Real-time VR or AR pipelines might use the decoupled parameters to update only the needed component instead of reprocessing the entire scene.
  • Combining the pipeline with external physics simulators could produce interactions whose lighting and material responses update automatically.

Load-bearing premise

The input data contain enough independent information to separate geometry, illumination, and material without extra constraints or multi-view capture.

What would settle it

Apply DR-GS to a glossy object that undergoes large non-rigid deformation under multiple new light directions and check whether the specular highlights remain physically consistent or revert to baked-in appearance.

Figures

Figures reproduced from arXiv: 2606.29379 by Jiaxin Li, Li Zhang, Tailin Wu, Tong Wu, Yi Wei.

Figure 1
Figure 1. Figure 1: DR-GS enables physically plausible rendering under deformations and lighting changes via decoupled geometry, lighting, and materials. The project page is available at https://jiaxinlia.github.io/DR-GS/ . Abstract. Gaussian splatting (GS) has garnered significant attention in VR/AR and digital content creation due to its explicit parameterization and efficient rendering capabilities. However, existing GS-ba… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of DR-GS. Our framework consists of three stages: (i) Static re￾construction: building initial 2D Gaussians and mesh from calibrated images, with optional mesh-based reinitialization; (ii) Parameters decoupling: generating material (albedo, metallic, roughness) and geometry (position, normal) maps via splatting while solving the rendering equation through MIS and ray tracing; (iii) Dynamic driving… view at source ↗
Figure 3
Figure 3. Figure 3: Qualitative comparisons on deformed reflective scenes [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative comparison of illumination decomposition and recon [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Relighting results on glossy objects. DR-GS more faithfully matches the GT specular highlights and reflection details while preserving material consistency, whereas IRGS often appears darker with weakened reflections and lost fine details. simulation and rendering: the former demonstrates gravity-driven soft-body de￾formation, where parameter changes markedly affect the Angel’s wings and the Horse’s legs (… view at source ↗
Figure 7
Figure 7. Figure 7: Relighting un￾der dim illumination. Consistent metallic sheen suggests DR-GS learns a more accurate base color [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: PBR results with material parameter editing. We present a comparison between DR-GS static reconstruction results and ground truth, along with rendered results after editing the decoupled material parameters. deform Particle -driven Mesh -driven Angel Bell Mutant NotEnrique Env. Teapot Vegeta [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: More results. It demonstrates reconstruction, geometric deformation, and PBR under various illumination conditions using estimated material parameters [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Ablation on mod￾eling of inter-reflection. w/o learnable 𝑑 GT Ours (full) Front view Back view w/o learnable 𝛼 [PITH_FULL_IMAGE:figures/full_fig_p014_11.png] view at source ↗
Figure 13
Figure 13. Figure 13: Effect of denoiser, MIS, and mesh acceleration on quality and run [PITH_FULL_IMAGE:figures/full_fig_p014_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Composed scenes. It demonstrates DR-GS’s ability to model occlusion￾induced shadows and reflections arising from geometric visibility effects. Complex non-manifold geometry. Unlike mesh-based methods, GS-based approaches better handle non-manifold topologies. On the TensoIR [22] (ficus) and Shelly [51] (khady, kitten, pug, wolly) datasets, covering foliage, hair, fur, and porous structures, DR-GS achieves… view at source ↗
Figure 15
Figure 15. Figure 15: Results of non-manifold geome￾try. It demonstrates DR-GS’s ability to deal with intricate structures. 5 Conclusion This paper presents DR-GS, a unified Gaussian framework for physically-based inverse rendering, relighting, and deformation-aware manipulation. By decou￾pling geometry, illumination, and materials, DR-GS addresses the unrealistic appearance of snapshots-based methods under geometric and light… view at source ↗
read the original abstract

Gaussian splatting (GS) has garnered significant attention in VR/AR and digital content creation due to its explicit parameterization and efficient rendering capabilities. However, existing GS-based methods for deformable objects face two key limitations: (i) illumination is erroneously baked into textures, causing physically inconsistent responses under dynamic deformations and lighting changes; (ii) snapshot-based reconstruction restricts post-reconstruction material editing. To address these challenges, we propose Deformable and Relightable GS (DR-GS), a unified Gaussian framework that integrates physically-based inverse rendering, relighting, and deformation-aware manipulation. Through explicitly disentangling geometry, illumination, and material representations, DR-GS overcomes the limitations of static snapshots, resolving unrealistic appearance under varying conditions while enabling post-reconstruction parameter editing. Extensive experiments show that DR-GS achieves leading visual quality across static reconstruction, dynamic deformation, and relighting, reliably preserving reflections and specular highlights on glossy surfaces. It further establishes a fully decoupled geometry-illumination-material pipeline, enabling high-quality 3D asset creation and comprehensive post-editing.

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

0 major / 2 minor

Summary. The paper proposes DR-GS, a unified 2D Gaussian splatting framework that integrates physically-based inverse rendering to explicitly disentangle geometry, illumination, and material representations for deformable objects. It addresses limitations in existing GS methods where illumination is baked into textures (causing inconsistent appearance under deformation and lighting changes) and snapshot-based reconstructions prevent post-editing of materials. The method enables relighting, deformation-aware manipulation, and parameter editing while preserving reflections and specular highlights; experiments claim leading visual quality across static reconstruction, dynamic deformation, and relighting tasks.

Significance. If the central claims hold, this advances explicit 3D representation methods by providing a physically consistent pipeline for dynamic scenes, which is valuable for VR/AR and digital content creation. The explicit disentanglement and post-reconstruction editing capabilities represent a meaningful step beyond static GS approaches, with potential for high-quality 3D asset pipelines.

minor comments (2)
  1. The abstract refers to 'extensive experiments' and 'leading visual quality' without specifying the datasets, baselines, or quantitative metrics used; this should be clarified in the introduction or experiments section for reproducibility.
  2. Notation for the disentangled representations (geometry, illumination, material) is introduced in the abstract but would benefit from explicit definitions and symbols in §3 or §4 to avoid ambiguity in the pipeline description.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their summary of our work and for recognizing its potential significance for dynamic scene representation and content creation pipelines. The recommendation is listed as uncertain, yet the report contains no specific major comments or requests for clarification. We therefore provide no point-by-point responses below. Should the referee have additional questions or concerns not captured in the provided report, we are happy to address them in a revised version or supplementary material.

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The abstract and description present DR-GS as integrating physically-based inverse rendering with explicit disentanglement of geometry, illumination, and material in a 2D Gaussian framework. No equations, fitted parameters, or self-citations are shown that reduce any claimed prediction or result to the inputs by construction. Central claims rest on experimental validation and post-reconstruction editing capabilities, which are independent of the method's internal definitions. This is the common honest finding for a methods paper whose core pipeline does not collapse into self-definition or fitted renaming.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract only; no information is provided on free parameters, axioms, or invented entities used in the method.

pith-pipeline@v0.9.1-grok · 5729 in / 1149 out tokens · 42931 ms · 2026-06-30T07:02:54.699706+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

64 extracted references · 6 canonical work pages

  1. [1]

    In: ICCV (2021) 3

    Boss, M., Braun, R., Jampani, V., Barron, J.T., Liu, C., Lensch, H.P.: Nerd: Neural reflectance decomposition from image collections. In: ICCV (2021) 3

    2021

  2. [2]

    In: NeurIPS (2022) 3

    Boss, M., Engelhardt, A., Kar, A., Li, Y., Sun, D., Barron, J.T., Lensch, H.P.A., Jampani, V.: Samurai: Shape and material from unconstrained real-world arbitrary image collections. In: NeurIPS (2022) 3

    2022

  3. [3]

    In: NeurIPS (2024) 3

    Cao, J., Guan, S., Ge, Y., Li, W., Yang, X., Ma, C.: NeuMA: Neural material adaptor for visual grounding of intrinsic dynamics. In: NeurIPS (2024) 3

    2024

  4. [4]

    In: ICLR (2025) 4

    Chen, H., Lin, Z., Zhang, J.: Gi-gs: Global illumination decomposition on gaussian splatting for inverse rendering. In: ICLR (2025) 4

    2025

  5. [5]

    Chen, Y., Gu, C., Jiang, J., Zhu, X., Zhang, L.: Periodic vibration gaussian: Dy- namicurbanscenereconstructionandreal-timerendering.arXiv:2311.18561(2023) 3

    work page arXiv 2023

  6. [6]

    ACM TOG (1982) 6

    Cook, R.L., Torrance, K.E.: A reflectance model for computer graphics. ACM TOG (1982) 6

    1982

  7. [7]

    In: CVPR (2025) 2, 3, 9, 10

    Feng, Y., Feng, X., Shang, Y., Jiang, Y., Yu, C., Zong, Z., Shao, T., Wu, H., Zhou, K., Jiang, C., Yang, Y.: Gaussian splashing: Unified particles for versatile motion synthesis and rendering. In: CVPR (2025) 2, 3, 9, 10

    2025

  8. [8]

    In: ECCV (2024) 4

    Gao, J., Gu, C., Lin, Y., Zhu, H., Cao, X., Zhang, L., Yao, Y.: Relightable 3d gaussian: Real-time point cloud relighting with brdf decomposition and ray tracing. In: ECCV (2024) 4

    2024

  9. [9]

    ACM TOG (2024) 2, 3, 5, 9

    Gao, L., Yang, J., Zhang, B., Sun, J., Yuan, Y., Fu, H., Lai, Y.K.: Real-time large-scale deformation of gaussian splatting. ACM TOG (2024) 2, 3, 5, 9

    2024

  10. [10]

    In: CVPR (2025) 2, 3, 9

    Gao,X.,Li,X.,Zhuang,Y.,Zhang,Q.,Hu,W.,Zhang,C.,Yao,Y.,Shan,Y.,Quan, L.: Mani-gs: Gaussian splatting manipulation with triangular mesh. In: CVPR (2025) 2, 3, 9

    2025

  11. [11]

    In: CVPR (2025) 2, 4, 5, 6, 9

    Gu, C., Wei, X., Zeng, Z., Yao, Y., Zhang, L.: Irgs: Inter-reflective gaussian splat- ting with 2d gaussian ray tracing. In: CVPR (2025) 2, 4, 5, 6, 9

    2025

  12. [12]

    In: CVPR (2025) 3

    Gu, C., Wei, X., Zhang, L., Zhu, X.: Tensoflow: Tensorial flow-based sampler for inverse rendering. In: CVPR (2025) 3

    2025

  13. [13]

    In: CVPR (2024) 2, 3, 9

    Guédon, A., Lepetit, V.: Sugar: Surface-aligned gaussian splatting for efficient 3d mesh reconstruction and high-quality mesh rendering. In: CVPR (2024) 2, 3, 9

    2024

  14. [14]

    In: CVPR (2025) 2

    Guo, M., Chiang, M.J.Y., Santesteban, I., Sarafianos, N., Chen, H.y., Halimi, O., Božič, A., Saito, S., Wu, J., Liu, C.K., Stuyck, T., Larionov, E.: Pgc: Physics-based gaussian cloth from a single pose. In: CVPR (2025) 2

    2025

  15. [15]

    In: MM (2024) 4

    Guo, Y., Bai, Y., Hu, L., Guo, Z., Liu, M., Cai, Y., Huang, T., Ma, L.: Prtgs: Pre- computed radiance transfer of gaussian splats for real-time high-quality relighting. In: MM (2024) 4

    2024

  16. [16]

    JCGT (2018) 2

    Heitz, E.: Sampling the ggx distribution of visible normals. JCGT (2018) 2

    2018

  17. [17]

    In: CVPR (2024) 2, 3

    Hu, L., Zhang, H., Zhang, Y., Zhou, B., Liu, B., Zhang, S., Nie, L.: Gaussianavatar: Towards realistic human avatar modeling from a single video via animatable 3d gaussians. In: CVPR (2024) 2, 3

    2024

  18. [18]

    In: SIGGRAPH (2024) 4

    Huang, B., Yu, Z., Chen, A., Geiger, A., Gao, S.: 2d gaussian splatting for geo- metrically accurate radiance fields. In: SIGGRAPH (2024) 4

    2024

  19. [19]

    In: AAAI (2025) 2, 3

    Huang, T., Zeng, Y., Li, H., Zuo, W., Lau, R.W.: Dreamphysics: Learning physical properties of dynamic 3d gaussians with video diffusion priors. In: AAAI (2025) 2, 3

    2025

  20. [20]

    In: CoRL (2024) 2 DR-GS: Physically-Based Deformable and Relightable 2D Gaussians 17

    Ji, M., Qiu, R.Z., Zou, X., Wang, X.: Graspsplats: Efficient manipulation with 3d feature splatting. In: CoRL (2024) 2 DR-GS: Physically-Based Deformable and Relightable 2D Gaussians 17

    2024

  21. [21]

    In: SIGGRAPH (2024) 2, 3

    Jiang, Y., Yu, C., Xie, T., Li, X., Feng, Y., Wang, H., Li, M., Lau, H., Gao, F., Yang, Y., Jiang, C.: Vr-gs: A physical dynamics-aware interactive gaussian splatting system in virtual reality. In: SIGGRAPH (2024) 2, 3

    2024

  22. [22]

    In: CVPR (2023) 8, 15

    Jin, H., Liu, I., Xu, P., Zhang, X., Han, S., Bi, S., Zhou, X., Xu, Z., Su, H.: Tensoir: Tensorial inverse rendering. In: CVPR (2023) 8, 15

    2023

  23. [23]

    In: Proceedings of the 13th Annual Confer- ence on Computer Graphics and Interactive Techniques (1986) 6

    Kajiya, J.T.: The rendering equation. In: Proceedings of the 13th Annual Confer- ence on Computer Graphics and Interactive Techniques (1986) 6

    1986

  24. [24]

    ACM TOG (2023) 3, 4, 7

    Kerbl, B., Kopanas, G., Leimkühler, T., Drettakis, G.: 3d gaussian splatting for real-time radiance field rendering. ACM TOG (2023) 3, 4, 7

    2023

  25. [25]

    TVCG (2025) 4

    Lai, S., Huang, L., Guo, J., Cheng, K., Pan, B., Long, X., Lyu, J., Lv, C., Guo, Y.: Glossygs: Inverse rendering of glossy objects with 3d gaussian splatting. TVCG (2025) 4

    2025

  26. [26]

    ArXiv (2025) 2, 3

    Li, M., Yao, S., Kai, C., Xie, Z., Chen, K., Jiang, Y.G.: Anigaussian: Animatable gaussian avatar with pose-guided deformation. ArXiv (2025) 2, 3

    2025

  27. [27]

    In: CVPR (2024) 3

    Li, Z., Chen, Z., Li, Z., Xu, Y.: Spacetime gaussian feature splatting for real-time dynamic view synthesis. In: CVPR (2024) 3

    2024

  28. [28]

    In: ICCV (2025) 2

    Li, Z., Yu, H.X., Liu, W., Yang, Y., Herrmann, C., Wetzstein, G., Wu, J.: Won- derplay: Dynamic 3d scene generation from a single image and actions. In: ICCV (2025) 2

    2025

  29. [29]

    In: CVPR (2024) 4

    Liang, Z., Zhang, Q., Feng, Y., Shan, Y., Jia, K.: Gs-ir: 3d gaussian splatting for inverse rendering. In: CVPR (2024) 4

    2024

  30. [30]

    Physics3d: Learning physical properties of 3d gaussians via video diffusion,

    Liu,F.,Wang,H.,Yao,S.,Zhang,S.,Zhou,J.,Duan,Y.:Physics3d:Learningphys- ical properties of 3d gaussians via video diffusion. arXiv preprint arXiv:2406.04338 (2024) 2, 3

    work page arXiv 2024

  31. [31]

    ACM Transactions on Graphics (ToG) (2023) 3, 8, 11, 12

    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) (2023) 3, 8, 11, 12

    2023

  32. [32]

    In: ECCV (2024) 2

    Lu, G., Zhang, S., Wang, Z., Liu, C., Lu, J., Tang, Y.: Manigaussian: Dynamic gaussian splatting for multi-task robotic manipulation. In: ECCV (2024) 2

    2024

  33. [33]

    ACM TOG (2025) 2, 3

    Luo, Z., Cui, Z., Luo, S., Chu, M., Li, M.: Vr-doh: Hands-on 3d modeling in virtual reality. ACM TOG (2025) 2, 3

    2025

  34. [34]

    In: ICCV (2025) 3

    Ma, Z., Jiang, J., Chen, Y., Zhang, L.: Béziergs: Dynamic urban scene reconstruc- tion with bézier curve gaussian splatting. In: ICCV (2025) 3

    2025

  35. [35]

    ACM Transactions on Graphics (TOG) (2010) 2

    Martin, S., Kaufmann, P., Botsch, M., Grinspun, E., Gross, M.: Unified simulation of elastic rods, shells, and solids. ACM Transactions on Graphics (TOG) (2010) 2

    2010

  36. [36]

    Commu- nications of the ACM (2021) 2

    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 ACM (2021) 2

    2021

  37. [37]

    ACM Transactions on Graphics (TOG)43(6), 1–19 (2024) 4

    Moenne-Loccoz, N.,Mirzaei, A.,Perel,O.,De Lutio,R., MartinezEsturo,J., State, G., Fidler, S., Sharp, N., Gojcic, Z.: 3d gaussian ray tracing: Fast tracing of particle scenes. ACM Transactions on Graphics (TOG)43(6), 1–19 (2024) 4

    2024

  38. [38]

    ACM TOG (2010) 4

    Parker, S.G., Bigler, J., Dietrich, A., Friedrich, H., Hoberock, J., Luebke, D., McAl- lister,D.,McGuire,M.,Morley,K.,Robison,A.,Stich,M.:Optix:ageneralpurpose ray tracing engine. ACM TOG (2010) 4

    2010

  39. [39]

    San Francisco, CA, USA, 3rd edn

    Pharr, M., Jakob, W., Humphreys, G.: Physically Based Rendering: From Theory to Implementation. San Francisco, CA, USA, 3rd edn. (2016) 2

    2016

  40. [40]

    In: CVPR (2024) 2, 3

    Qian, S., Kirschstein, T., Schoneveld, L., Davoli, D., Giebenhain, S., Nießner, M.: Gaussianavatars: Photorealistic head avatars with rigged 3d gaussians. In: CVPR (2024) 2, 3

    2024

  41. [41]

    In: ECCV (2024) 3 18 J

    Qiu, R.Z., Yang, G., Zeng, W., Wang, X.: Language-driven physics-based scene synthesis and editing via feature splatting. In: ECCV (2024) 3 18 J. Li et al

    2024

  42. [42]

    In: Proceed- ings of High Performance Graphics (2017) 2, 8

    Schied, C., Kaplanyan, A., Wyman, C., Patney, A., Chaitanya, C.R.A., Burgess, J., Liu, S., Dachsbacher, C., Lefohn, A., Salvi, M.: Spatiotemporal variance-guided filtering: real-time reconstruction for path-traced global illumination. In: Proceed- ings of High Performance Graphics (2017) 2, 8

    2017

  43. [43]

    ACM TOG (2023) 7

    Shen, T., Munkberg, J., Hasselgren, J., Yin, K., Wang, Z., Chen, W., Gojcic, Z., Fidler, S., Sharp, N., Gao, J.: Flexible isosurface extraction for gradient-based mesh optimization. ACM TOG (2023) 7

    2023

  44. [44]

    T-PAMI (2025) 4

    Shi, Y., Wu, Y., Wu, C., Liu, X., Zhao, C., Feng, H., Zhang, J., Zhou, B., Ding, E., Wang, J.: Gir: 3d gaussian inverse rendering for relightable scene factorization. T-PAMI (2025) 4

    2025

  45. [45]

    In: 8th Annual Conference on Robot Learning (2024) 2

    Shorinwa, O., Tucker, J., Smith, A., Swann, A., Chen, T., Firoozi, R., Kennedy, M.D., Schwager, M.: Splat-mover: Multi-stage, open-vocabulary robotic manipula- tion via editable gaussian splatting. In: 8th Annual Conference on Robot Learning (2024) 2

    2024

  46. [46]

    In: CVPR (2021) 3

    Srinivasan, P.P., Deng, B., Zhang, X., Tancik, M., Mildenhall, B., Barron, J.T.: Nerv: Neural reflectance and visibility fields for relighting and view synthesis. In: CVPR (2021) 3

    2021

  47. [47]

    In: CVPR (2025) 4

    Sun, H., Gao, Y., Xie, J., Yang, J., Wang, B.: Svg-ir: Spatially-varying gaussian splatting for inverse rendering. In: CVPR (2025) 4

    2025

  48. [48]

    In: Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques (1995) 8

    Veach, E., Guibas, L.J.: Optimally combining sampling techniques for monte carlo rendering. In: Proceedings of the 22nd Annual Conference on Computer Graphics and Interactive Techniques (1995) 8

    1995

  49. [49]

    arXiv preprint arXiv:2402.01459 (2024) 3

    Waczyńska, J., Borycki, P., Tadeja, S., Tabor, J., Spurek, P.: Games: Mesh-based adapting and modification of gaussian splatting. arXiv preprint arXiv:2402.01459 (2024) 3

    work page arXiv 2024

  50. [50]

    IEEE Transactions on Image Processing (2004) 9

    Wang, Z., Bovik, A., Sheikh, H., Simoncelli, E.: Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing (2004) 9

    2004

  51. [51]

    ACM Trans

    Wang, Z., Shen, T., Nimier-David, M., Sharp, N., Gao, J., Keller, A., Fidler, S., Müller, T., Gojcic, Z.: Adaptive shells for efficient neural radiance field rendering. ACM Trans. Graph.42(6) (2023).https://doi.org/10.1145/3618390,https: //doi.org/10.1145/361839015

    work page doi:10.1145/3618390 2023

  52. [52]

    In: CVPR (2024) 3

    Wu, G., Yi, T., Fang, J., Xie, L., Zhang, X., Wei, W., Liu, W., Tian, Q., Wang, X.: 4d gaussian splatting for real-time dynamic scene rendering. In: CVPR (2024) 3

    2024

  53. [53]

    In: CVPR (2024) 2, 3, 9, 10

    Xie, T., Zong, Z., Qiu, Y., Li, X., Feng, Y., Yang, Y., Jiang, C.: Physgaussian: Physics-integrated 3d gaussians for generative dynamics. In: CVPR (2024) 2, 3, 9, 10

    2024

  54. [54]

    In: ICLR (2024) 3

    Yang, Z., Yang, H., Pan, Z., Zhang, L.: Real-time photorealistic dynamic scene representation and rendering with 4d gaussian splatting. In: ICLR (2024) 3

    2024

  55. [55]

    In: CVPR (2024) 3

    Yang, Z., Gao, X., Zhou, W., Jiao, S., Zhang, Y., Jin, X.: Deformable 3d gaussians for high-fidelity monocular dynamic scene reconstruction. In: CVPR (2024) 3

    2024

  56. [56]

    In: ICLR (2025) 4, 5, 9

    Yao, Y., Zeng, Z., Gu, C., Zhu, X., Zhang, L.: Reflective gaussian splatting. In: ICLR (2025) 4, 5, 9

    2025

  57. [57]

    Real2render2real: Scalingrobotdatawithoutdynamicssimulation or robot hardware.arXiv preprint arXiv:2505.09601, 2025

    Yu, J., Fu, L., Huang, H., El-Refai, K., Ambrus, R.A., Cheng, R., Irshad, M.Z., Goldberg, K.: Real2render2real: Scaling robot data without dynamics simulation or robot hardware. arXiv preprint arXiv:2505.09601 (2025) 2

    work page arXiv 2025

  58. [58]

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

    Zhang, J., Yao, Y., Li, S., Liu, J., Fang, T., McKinnon, D., Tsin, Y., Quan, L.: Neilf++: Inter-reflectable light fields for geometry and material estimation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 3601–3610 (2023) 3 DR-GS: Physically-Based Deformable and Relightable 2D Gaussians 19

    2023

  59. [59]

    In: CVPR (2018) 9

    Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: CVPR (2018) 9

    2018

  60. [60]

    In: ECCV (2024) 3

    Zhang, T., Yu, H.X., Wu, R., Feng, B.Y., Zheng, C., Snavely, N., Wu, J., Freeman, W.T.: PhysDreamer: Physics-based interaction with 3d objects via video genera- tion. In: ECCV (2024) 3

    2024

  61. [61]

    Efficient physics simulation for 3d scenes via mllm-guided gaussian splatting,

    Zhao, H., Wang, H., Zhao, X., Fei, H., Wang, H., Long, C., Zou, H.: Efficient physics simulation for 3d scenes via mllm-guided gaussian splatting. arXiv preprint arXiv:2411.12789 (2024) 3

    work page arXiv 2024

  62. [62]

    IEEE Robotics and Automation Letters (2024) 2

    Zheng, Y., Chen, X., Zheng, Y., Gu, S., Yang, R., Jin, B., Li, P., Zhong, C., Wang, Z., Liu, L., et al.: Gaussiangrasper: 3d language gaussian splatting for open- vocabulary robotic grasping. IEEE Robotics and Automation Letters (2024) 2

    2024

  63. [63]

    NeurIPS (2024) 3

    Zhu, T., Chen, Z., Gao, J., Yan, Y., Yang, X.: Multi-times monte carlo rendering for inter-reflection reconstruction. NeurIPS (2024) 3

    2024

  64. [64]

    In: ACM TOG (2025) 9

    Zhu, Z.L., Wang, B., Yang, J.: Gs-ror2: Bidirectional-guided 3dgs and sdf for re- flective object relighting and reconstruction. In: ACM TOG (2025) 9

    2025