Pith. sign in

REVIEW 3 major objections 2 minor 48 references

Laplacian Analysis Meets Dynamics Modelling: Gaussian Splatting for 4D Reconstruction

T0 review · 3 major / 2 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper proposes a hybrid explicit–implicit Gaussian Splatting framework that uses spectral-aware Laplacian encoding, an enhanced Gaussian dynamics attribute, and KDTree-based splitting to reconstruct dynamic 4D scenes with better fidelit

desk verdict Plausible dynamic 3DGS framework with a solid problem framing, but the abstract's central spectral-conflict claim is unsupported and the actual evidence is invisible. read the letter →

arxiv 2508.04966 v1 pith:JA7AX4UU submitted 2025-08-07 cs.GR cs.CVcs.MM

Yifan Zhou , Beizhen Zhao , Pengcheng Wu , Hao Wang This is my paper · ORCID
classification cs.GRcs.CVcs.MM
keywords 4DreconstructionGaussianSplattingLaplacianencodingdynamicsceneshashKDTreesplittingspectralanalysisexplicit-implicitrepresentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Dynamic 3D Gaussian Splatting (3DGS) methods tend to either over-smooth motion or suffer from feature collisions. The paper argues the root cause is a spectral conflict between preserving motion detail and maintaining deformation consistency. It proposes a hybrid explicit–implicit framework that separates these frequencies by merging hash encoding with a Laplacian module, adds a Gaussian dynamics attribute to compensate for photometric distortions, and uses a KDTree-based split strategy to focus optimization on dynamic regions. The authors report state-of-the-art reconstruction fidelity on complex dynamic scenes.

What carries the argument

The load-bearing piece is the Laplacian module added to the hash encoder: a spectral filter bank that separates high-frequency motion detail from low-frequency deformation, letting the network control each without collision. The other two components are the enhanced Gaussian dynamics attribute, which absorbs photometric changes tied to deformation, and the KDTree-based split, which decides where to densify Gaussians by querying dynamic regions efficiently.

What would settle it

Run both the full model and a hash-only ablation on a scene with a known ground-truth deformation field and independently measured frequency content of motion and geometry. If the hash-only baseline already matches the full model's fidelity, or if the frequency gap between motion and deformation does not correlate with the quality gap, the spectral-conflict premise would not be supported.

Watch

Extended reading notes

Core claim

The central claim is that dynamic scenes can be reconstructed more faithfully by treating the motion/detail conflict as a spectral separation problem. Concretely, the paper introduces a spectral-aware Laplacian encoding architecture that merges Hash encoding with a Laplacian-based module to give independent frequency control over motion detail and deformation consistency, an enhanced Gaussian dynamics attribute that corrects photometric distortion caused by geometric deformation, and an adaptive split strategy guided by KDTree primitive control. The authors state that these components together achieve state-of-the-art reconstruction fidelity in complex dynamic scenes.

Load-bearing premise

The assumption that matters is that adding a Laplacian module to hash encoding genuinely separates motion detail from deformation consistency in frequency space; the abstract asserts this conflict and its resolution without derivation or dedicated evidence.

Editorial extensions

If this is right

  • If the spectral conflict is real and the Laplacian separation resolves it, dynamic 3DGS can render fast-moving, thin structures without either blurring them or flickering.
  • The enhanced dynamics attribute means photometric compensation is handled inside the Gaussian representation, so lighting changes from deformation need not be baked into geometry.
  • KDTree-guided splitting could make optimization more sample-efficient in scenes where motion is spatially concentrated.
  • The method reframes dynamic reconstruction as a frequency-management problem, suggesting that other explicit–implicit hybrid representations could benefit from spectral control.

Reading between the lines

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

  • A testable consequence not stated in the abstract: the benefit of the Laplacian module should grow as the spectral overlap between motion and geometry increases; scenes with slow, rigid motion would show little difference from a hash-only baseline.
  • The paper implies that the spectral conflict is inherent to the representation rather than to the optimizer; one could probe this by training the baseline longer or with different schedules and checking whether the conflict disappears.
  • The KDTree split could also be applied to static scenes with heterogeneous detail density, a transfer the paper does not claim.
  • If the claim holds, a practical extension is to make the Laplacian filter bandwidth a learned or scene-adaptive parameter instead of a fixed design choice.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 2 minor

Summary. The paper proposes a 4D reconstruction framework built on 3D Gaussian Splatting, with three claimed innovations: a spectral-aware Laplacian encoding architecture that merges Hash encoding with a Laplacian-based module for frequency-controlled motion modeling, an enhanced Gaussian dynamics attribute to compensate for photometric distortion, and a KDTree-guided adaptive Gaussian split strategy. The abstract asserts that existing dynamic 3DGS methods suffer from an inherent spectral conflict between motion detail preservation and deformation consistency, and that the proposed framework resolves this conflict, achieving state-of-the-art reconstruction fidelity.

Significance. If substantiated, the paper would address a genuine limitation of dynamic 3D Gaussian Splatting: the trade-off between preserving high-frequency motion details and maintaining coherent deformation across time. The hybrid explicit-implicit architecture and the proposed spectral decomposition could be a meaningful step forward. However, the available manuscript is only the abstract; it contains no derivations, architectural details, ablations, or quantitative comparisons. The central mechanistic claim—that a Laplacian module resolves a spectral conflict—is therefore entirely unsupported in the current material, and the state-of-the-art claim cannot be checked. The significance is potentially high but currently unverified.

major comments (3)
  1. [Abstract] The central premise of the paper is that an 'inherent spectral conflict' exists between preserving motion details and maintaining deformation consistency, and that this conflict arises from low-rank decomposition and high-dimensional grid sampling. This is asserted without formal definition or derivation. No spectral decomposition is specified, no argument is given for why the two terms occupy conflicting frequency bands, and no mechanism is described for how the Laplacian module separates them. Since the claimed advantage rests on this mechanism, this is load-bearing and requires either a formal analysis or an ablation isolating the Laplacian module's contribution.
  2. [Abstract] The claim of 'state-of-the-art performance' and 'better reconstruction fidelity' is not accompanied by any quantitative evidence. No datasets, metrics, baselines, or error bars are reported. The sentence 'Through extensive experiments...' is an assertion, not a result. Even for an abstract, a representative quantitative comparison or a reference to a results table would be necessary to support the central empirical claim.
  3. [Abstract] The two remaining innovations—'enhanced Gaussian dynamics attribute' and 'KDTree-based split strategy'—are named but not explained. It is not stated what information the dynamics attribute encodes, how it compensates for photometric distortion, how the KDTree-based split differs from existing adaptive density control, or why it is more efficient. Without these details, the novelty and correctness of these components cannot be evaluated.
minor comments (2)
  1. [Abstract] Minor language issue: 'at different frequency' should be 'at different frequencies'.
  2. [Abstract] The phrase 'flexible frequency motion control' is vague. Clarifying the intended frequency bands and how the Laplacian module adjusts them would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No identifiable circularity: the abstract asserts an unsupported spectral-conflict mechanism and an empirical SOTA claim, but no derivation step reduces to the paper's own inputs.

full rationale

The paper as provided consists only of the abstract and a brief method description; no derivation chain, equations, fitted parameters, or self-citations are exposed. The central premise—that dynamic 3DGS suffers from an 'inherent spectral conflict' between motion detail preservation and deformation consistency—is asserted rather than derived, and the proposed Laplacian-hash hybrid is described without a formal frequency decomposition or proof of separation. These are unsupported empirical/mechanistic claims, not circular ones: the abstract does not define the method's output in terms of its input, rename a fitted quantity as a prediction, or rely on a load-bearing self-citation. The SOTA performance claim is an external benchmark comparison and is therefore falsifiable rather than tautological. Under the hard rule that circularity must be exhibited by quoting a specific reduction, no such reduction is present. The appropriate finding is therefore no significant circularity, with the caveat that the spectral-conflict justification is under-derived rather than self-referential.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

Only the abstract was available, so the axiom ledger is minimal and based on the abstract's own assertions. No free parameters or invented physical/theoretical entities are stated.

assumptions (2)
  • domain assumption Existing dynamic 3DGS methods suffer from over-smoothing due to low-rank decomposition or feature collision from high-dimensional grid sampling.
    This characterizes prior work but is asserted in the abstract without supporting analysis or citations.
  • domain assumption Spectral conflicts between preserving motion details and maintaining deformation consistency can be addressed by adding a Laplacian encoding module to a hash grid.
    The core design premise is that frequency separation through a Laplacian module yields the claimed benefit; this is not derived or shown in the abstract.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Laplacian Analysis Meets Dynamics Modelling: Gaussian Splatting for 4D Reconstruction." pith.science (2026). https://pith.science/paper/JA7AX4UU

@misc{pith2026250804966,
  author       = {Pith},
  title        = {Pith review of: Laplacian Analysis Meets Dynamics Modelling: Gaussian Splatting for 4D Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JA7AX4UU}},
  note         = {Machine review of arXiv:2508.04966}
}
read the original abstract

While 3D Gaussian Splatting (3DGS) excels in static scene modeling, its extension to dynamic scenes introduces significant challenges. Existing dynamic 3DGS methods suffer from either over-smoothing due to low-rank decomposition or feature collision from high-dimensional grid sampling. This is because of the inherent spectral conflicts between preserving motion details and maintaining deformation consistency at different frequency. To address these challenges, we propose a novel dynamic 3DGS framework with hybrid explicit-implicit functions. Our approach contains three key innovations: a spectral-aware Laplacian encoding architecture which merges Hash encoding and Laplacian-based module for flexible frequency motion control, an enhanced Gaussian dynamics attribute that compensates for photometric distortions caused by geometric deformation, and an adaptive Gaussian split strategy guided by KDTree-based primitive control to efficiently query and optimize dynamic areas. Through extensive experiments, our method demonstrates state-of-the-art performance in reconstructing complex dynamic scenes, achieving better reconstruction fidelity.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 33 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Bae, J.; Kim, S.; Yun, Y.; Lee, H.; Bang, G.; and Uh, Y. 2024. Per-gaussian embedding-based deformation for deformable 3d gaussian splatting. In European Conference on Computer Vision, 321--335. Springer

    work page 2024

  4. [4]

    T.; Mildenhall, B.; Tancik, M.; Hedman, P.; Martin-Brualla, R.; and Srinivasan, P

    Barron, J. T.; Mildenhall, B.; Tancik, M.; Hedman, P.; Martin-Brualla, R.; and Srinivasan, P. P. 2021. Mip-nerf: A multiscale representation for anti-aliasing neural radiance fields. In Proceedings of the IEEE/CVF international conference on computer vision, 5855--5864

    work page 2021

  5. [5]

    T.; Mildenhall, B.; et al

    Barron, J. T.; Mildenhall, B.; et al. 2023. Zip-nerf: Anti-aliased grid-based neural radiance fields. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 19697--19705

    work page 2023

  6. [6]

    Cai, H.; Feng, W.; Feng, X.; Wang, Y.; and Zhang, J. 2022. Neural surface reconstruction of dynamic scenes with monocular rgb-d camera. Advances in Neural Information Processing Systems, 35: 967--981

    work page 2022

  7. [7]

    Cao, A.; and Johnson, J. 2023. Hexplane: A fast representation for dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 130--141

    2023

  8. [8]

    R.; Lin, C

    Chan, E. R.; Lin, C. Z.; Chan, M. A.; Nagano, K.; Pan, B.; De Mello, S.; Gallo, O.; Guibas, L. J.; Tremblay, J.; Khamis, S.; et al. 2022. Efficient geometry-aware 3d generative adversarial networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 16123--16133

    work page 2022

Show all 48 references
  1. [9]

    S.; and Anandkumar, A

    Choe, J.; Choy, C.; Park, J.; Kweon, I. S.; and Anandkumar, A. 2023. Spacetime surface regularization for neural dynamic scene reconstruction. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 17871--17881

    2023

  2. [10]

    B.; and Wu, J

    Du, Y.; Zhang, Y.; Yu, H.-X.; Tenenbaum, J. B.; and Wu, J. 2021. Neural radiance flow for 4d view synthesis and video processing. In 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 14304--14314. IEEE Computer Society

    2021

  3. [11]

    Duan, Y.; Wei, F.; Dai, Q.; He, Y.; Chen, W.; and Chen, B. 2024. 4d gaussian splatting: Towards efficient novel view synthesis for dynamic scenes. arXiv e-prints, arXiv--2402

    2024

  4. [12]

    P.; Mandi, Z.; Yao, Y.; Liu, J.-W.; Shou, M

    Duisterhof, B. P.; Mandi, Z.; Yao, Y.; Liu, J.-W.; Shou, M. Z.; Song, S.; and Ichnowski, J. 2023. Md-splatting: Learning metric deformation from 4d gaussians in highly deformable scenes. arXiv preprint arXiv:2312.00583, 2(3)

    2023 arXiv

  5. [13]

    Fang, J.; Yi, T.; Wang, X.; Xie, L.; Zhang, X.; Liu, W.; Nie ner, M.; and Tian, Q. 2022. Fast dynamic radiance fields with time-aware neural voxels. In SIGGRAPH Asia 2022 Conference Papers, 1--9

    2022

  6. [14]

    R.; Recht, B.; and Kanazawa, A

    Fridovich-Keil, S.; Meanti, G.; Warburg, F. R.; Recht, B.; and Kanazawa, A. 2023. K-planes: Explicit radiance fields in space, time, and appearance. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12479--12488

    2023

  7. [15]

    Gao, C.; Saraf, A.; Kopf, J.; and Huang, J.-B. 2021. Dynamic view synthesis from dynamic monocular video. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 5712--5721

    2021

  8. [16]

    Huang, Y.-H.; Sun, Y.-T.; Yang, Z.; Lyu, X.; Cao, Y.-P.; and Qi, X. 2024. Sc-gs: Sparse-controlled gaussian splatting for editable dynamic scenes. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 4220--4230

    2024

  9. [17]

    Kerbl, B.; Kopanas, G.; Leimk \"u hler, T.; and Drettakis, G. 2023. 3D Gaussian splatting for real-time radiance field rendering. ACM Trans. Graph., 42(4): 139--1

    2023

  10. [18]

    Kratimenos, A.; Lei, J.; and Daniilidis, K. 2024. Dynmf: Neural motion factorization for real-time dynamic view synthesis with 3d gaussian splatting. In European Conference on Computer Vision, 252--269. Springer

    2024

  11. [19]

    Y.; Cheong, W.-S.; Oh, J.; and Kim, M

    Kwak, S.; Kim, J.; Jeong, J. Y.; Cheong, W.-S.; Oh, J.; and Kim, M. 2025. MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni...

    2025

  12. [20]

    Li, T.; Slavcheva, M.; et al. 2022. Neural 3d video synthesis from multi-view video. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 5521--5531

    2022

  13. [21]

    Li, Z.; Chen, Z.; Li, Z.; and Xu, Y. 2024. Spacetime gaussian feature splatting for real-time dynamic view synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8508--8520

    2024

  14. [22]

    Liang, Y.; Khan, N.; Li, Z.; Nguyen-Phuoc, T.; Lanman, D.; Tompkin, J.; and Xiao, L. 2023. Gaufre: Gaussian deformation fields for real-time dynamic novel view synthesis. arXiv preprint arXiv:2312.11458

    2023 arXiv

  15. [23]

    Liang, Y.; Laidlaw, E.; et al. 2023. Semantic attention flow fields for monocular dynamic scene decomposition. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 21797--21806

    2023

  16. [24]

    Lin, Y.; Dai, Z.; Zhu, S.; and Yao, Y. 2024. Gaussian-flow: 4d reconstruction with dynamic 3d gaussian particle. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21136--21145

    2024

  17. [25]

    Liu, Y.-L.; Gao, C.; Meuleman, A.; Tseng, H.-Y.; Saraf, A.; Kim, C.; Chuang, Y.-Y.; Kopf, J.; and Huang, J.-B. 2023. Robust dynamic radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 13--23

    2023

  18. [26]

    Lu, Z.; Guo, X.; Hui, L.; Chen, T.; Yang, M.; Tang, X.; Zhu, F.; and Dai, Y. 2024. 3d geometry-aware deformable gaussian splatting for dynamic view synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8900--8910

    2024

  19. [27]

    Luiten, J.; Kopanas, G.; Leibe, B.; and Ramanan, D. 2024. Dynamic 3d gaussians: Tracking by persistent dynamic view synthesis. In 2024 International Conference on 3D Vision (3DV), 800--809. IEEE

    2024

  20. [28]

    S.; Barron, J

    Martin-Brualla, R.; Radwan, N.; Sajjadi, M. S.; Barron, J. T.; Dosovitskiy, A.; and Duckworth, D. 2021. Nerf in the wild: Neural radiance fields for unconstrained photo collections. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 7210--7219

    2021

  21. [29]

    P.; Tancik, M.; Barron, J

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

    2021

  22. [30]

    M \"u ller, T.; Evans, A.; Schied, C.; and Keller, A. 2022. Instant neural graphics primitives with a multiresolution hash encoding. ACM transactions on graphics (TOG), 41(4): 1--15

    2022

  23. [31]

    V.; Bello, J

    Park, J.; Bui, M.-Q. V.; Bello, J. L. G.; Moon, J.; Oh, J.; and Kim, M. 2024. SplineGS: Robust Motion-Adaptive Spline for Real-Time Dynamic 3D Gaussians from Monocular Video. arXiv preprint arXiv:2412.09982

    2024 arXiv

  24. [32]

    T.; Bouaziz, S.; Goldman, D

    Park, K.; Sinha, U.; Barron, J. T.; Bouaziz, S.; Goldman, D. B.; Seitz, S. M.; and Martin-Brualla, R. 2021 a . Nerfies: Deformable neural radiance fields. In Proceedings of the IEEE/CVF international conference on computer vision, 5865--5874

    2021

  25. [33]

    T.; Bouaziz, S.; Goldman, D

    Park, K.; Sinha, U.; Hedman, P.; Barron, J. T.; Bouaziz, S.; Goldman, D. B.; Martin-Brualla, R.; and Seitz, S. M. 2021 b . Hypernerf: A higher-dimensional representation for topologically varying neural radiance fields. arXiv preprint arXiv:2106.13228

    2021 arXiv

  26. [34]

    Pumarola, A.; Corona, E.; Pons-Moll, G.; and Moreno-Noguer, F. 2021. D-nerf: Neural radiance fields for dynamic scenes. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 10318--10327

    2021

  27. [35]

    Shao, R.; Zheng, Z.; Tu, H.; Liu, B.; Zhang, H.; and Liu, Y. 2023. Tensor4d: Efficient neural 4d decomposition for high-fidelity dynamic reconstruction and rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16632--16642

    2023

  28. [36]

    P.; Barron, J

    Tancik, M.; Casser, V.; Yan, X.; Pradhan, S.; Mildenhall, B.; Srinivasan, P. P.; Barron, J. T.; and Kretzschmar, H. 2022. Block-nerf: Scalable large scene neural view synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 8248--8258

    2022

  29. [37]

    Tretschk, E.; Tewari, A.; et al. 2021. Non-rigid neural radiance fields: Reconstruction and novel view synthesis of a dynamic scene from monocular video. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 12959--12970

    2021

  30. [38]

    E.; Jeni, L

    Wang, C.; MacDonald, L. E.; Jeni, L. A.; and Lucey, S. 2023. Flow supervision for deformable nerf. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21128--21137

    2023

  31. [39]

    C.; Sheikh, H

    Wang, Z.; Bovik, A. C.; Sheikh, H. R.; and Simoncelli, E. P. 2004. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing, 13(4): 600--612

    2004

  32. [40]

    Wu, G.; Yi, T.; Fang, J.; Xie, L.; Zhang, X.; Wei, W.; Liu, W.; Tian, Q.; and Wang, X. 2024. 4d gaussian splatting for real-time dynamic scene rendering. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 20310--20320

    2024

  33. [41]

    Xie, T.; Zong, Z.; et al. 2024. Physgaussian: Physics-integrated 3d gaussians for generative dynamics. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4389--4398

    2024

  34. [42]

    Xu, J.; Fan, Z.; Yang, J.; and Xie, J. 2024. Grid4D: 4D Decomposed Hash Encoding for High-Fidelity Dynamic Gaussian Splatting. arXiv preprint arXiv:2410.20815

    2024 arXiv

  35. [43]

    Yang, G.; Vo, M.; et al. 2022. Banmo: Building animatable 3d neural models from many casual videos. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2863--2873

    2022

  36. [44]

    Yang, Z.; Gao, X.; Zhou, W.; Jiao, S.; Zhang, Y.; and Jin, X. 2024. Deformable 3d gaussians for high-fidelity monocular dynamic scene reconstruction. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 20331--20341

    2024

  37. [45]

    Yang, Z.; Yang, H.; Pan, Z.; and Zhang, L. 2023. Real-time photorealistic dynamic scene representation and rendering with 4d gaussian splatting. arXiv preprint arXiv:2310.10642

    2023 arXiv

  38. [46]

    Yoo, J.-C.; and Han, T. H. 2009. Fast normalized cross-correlation. Circuits, systems and signal processing, 28: 819--843

    2009

  39. [47]

    \'A .; Niinuma, K.; and Jeni, L

    Yu, H.; Julin, J.; Milacski, Z. \'A .; Niinuma, K.; and Jeni, L. A. 2024. Cogs: Controllable gaussian splatting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21624--21633

    2024

  40. [48]

    Zhu, R.; Liang, Y.; Chang, H.; Deng, J.; Lu, J.; Yang, W.; Zhang, T.; and Zhang, Y. 2024. Motiongs: Exploring explicit motion guidance for deformable 3d gaussian splatting. Advances in Neural Information Processing Systems, 37: 101790--101817

    2024

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.