Pith. sign in

REVIEW 3 major objections 6 minor 42 references

ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper claims that deforming a sparse set of anchors instead of every Gaussian yields 4D scenes at 4-6 MB and 100+ FPS with little quality loss.

desk verdict Solid system paper on anchor-driven 4DGS compression with a genuine storage win, but the headline FPS claim needs controlled benchmarking before I'd trust the speed numbers. read the letter →

arxiv 2505.08196 v1 pith:WYQ2VLJA submitted 2025-05-13 cs.CV

He Huang , Qi Yang , Mufan Liu , Yiling Xu , Zhu Li This is my paper · ORCID
classification cs.CV
keywords 4DGaussiansplattingdynamicscenereconstructionanchor-basedrepresentationcoarse-to-finedeformationrate-distortionoptimizationentropymodelviewsynthesiscompression
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

ADC-GS claims that dynamic scenes can be reconstructed and rendered far more cheaply than existing 4D Gaussian splatting methods by replacing per-Gaussian deformation with anchor-driven deformation. The paper organizes Gaussian primitives under sparse anchors in a canonical space, deforms only the anchors at each frame, and refines per-primitive opacity and color in a fine stage. A multi-dimension entropy model estimates bitrate and trains the whole representation under a rate-distortion loss, shrinking models to about 4-6 MB while rendering at 101-135 FPS on the tested benchmarks. The payoff, if true, is that high-quality dynamic free-viewpoint video becomes practical to store, transmit, and play in real time.

What carries the argument

The central object is the anchor representation: a sparse set of anchors, each associated with K neural Gaussian primitives and stored as a compact residual form with a reference feature, a residual feature, and explicit Gaussian attributes. The coarse-to-fine deformation pipeline uses a tiny MLP to deform anchor attributes and a second MLP to refine per-primitive opacity and color, so the number of deformations per frame depends on the anchor count rather than the full Gaussian count. The multi-dimension entropy model estimates bitrate from adaptively quantized features using a hyperprior and contextual chunk-wise prediction, enabling rate-distortion optimization. The temporal-significance refinement grows and prunes anchors based on gradient-weighted rendering weights and accumulated opacity, addressing under- and over-reconstruction in dynamic scenes.

What would settle it

Run the per-Gaussian deformation baseline and ADC-GS on the same GPU, same evaluation frames, and same timing protocol, then check whether ADC-GS is still several times faster at matched model size and quality; if the gap disappears, the central speed claim is false.

Watch

Extended reading notes

Core claim

ADC-GS establishes that the redundancy among neighboring Gaussian primitives in 4D Gaussian splatting is large enough that motion can be modeled by deforming a sparse set of anchors rather than every Gaussian. In the canonical space at time zero, each anchor stores a shared reference feature, per-primitive residual features, and explicit position, covariance, and color attributes, from which K Gaussian primitives are generated. At each target frame, a coarse stage deforms the anchors and thereby updates all associated primitives, while a fine stage adjusts opacity and color per primitive. Combined with temporal-significance-based anchor growing and pruning and rate-distortion optimization, this yields models of roughly 4-6 MB with PSNR within about 0.3-0.6 dB of the strongest deformable baseline while rendering 3-8 times faster.

Load-bearing premise

The speed advantage rests on comparing ADC-GS's frame rates against baseline frame rates that may have been measured on different hardware or with different evaluation settings, since the paper does not report the baselines' GPUs.

Editorial extensions

If this is right

  • Models shrink to roughly 4-6 MB per scene, a 10x or larger reduction versus deformation-based 4DGS, making dynamic scenes practical to store and transmit.
  • Rendering reaches 101-135 FPS on the tested multi-view datasets, fast enough for interactive playback on a single consumer GPU.
  • Varying the Lagrange multiplier traces a rate-distortion curve, so the same representation covers a range of storage and quality trade-offs.
  • Quality stays within roughly 0.3-0.6 dB PSNR of the strongest per-Gaussian deformation baseline, so the compression does not come at a large fidelity cost.

Reading between the lines

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

  • If anchor-driven deformation generalizes, the same structure could be applied to monocular or sparse-view dynamic scenes, where per-Gaussian deformation cost is a known bottleneck.
  • The temporal-significance weighting used for anchor growing could be repurposed as a saliency prior for allocating bits to motion-heavy regions in dynamic scene compression.
  • Because anchors are shared across all frames, the method suggests model size may stay roughly constant as sequence length grows; a direct test on longer captures would confirm whether storage scales sublinearly.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper presents ADC-GS, a 4D Gaussian splatting representation for dynamic scenes. Instead of deforming each Gaussian independently, ADC-GS organizes Gaussian primitives under a sparse set of anchors in a canonical space. Each anchor predicts K Gaussian primitives; a coarse-to-fine deformation pipeline first deforms the anchor's explicit attributes and then refines per-primitive opacity and color. A multi-dimension entropy model is used in a rate-distortion optimization to compress anchor features, and temporal-significance-based anchor growing and pruning is introduced. Experiments on HyperNeRF and Neu3D report model sizes of roughly 4–7 MB, rendering speeds of 101–135 FPS, and PSNR values within about 0.3–0.6 dB of the best deformation-based baseline (E-D3DGS). The paper includes ablations, per-scene results, and an analysis of bitstream composition.

Significance. If the reported figures are reproducible, ADC-GS would constitute a practically useful step for compact and fast dynamic scene rendering. The anchor-driven deformation idea is simple and plausible, and the component-wise ablation supports the claim that each proposed module contributes to the final trade-off. The code is released, which aids reproducibility. However, the two headline claims—the 300–800% rendering-speed advantage and the statement that quality is preserved without compromise—are not fully supported by the evidence as presented. The speed comparison lacks controlled baseline measurements on comparable hardware, and the LPIPS numbers are consistently worse than E-D3DGS on both datasets. These issues are load-bearing for the abstract and conclusions, so they need to be resolved before the central claims can be accepted.

major comments (3)
  1. [Section 4.1, Tables 1 and 2] The FPS comparison for baselines lacks hardware and measurement details. The paper specifies that the authors' model is trained on a single NVIDIA RTX 3090, but it does not state the GPU, rendering resolution, or measurement protocol used for the baseline FPS values (e.g., E-D3DGS 26 FPS on HyperNeRF and 42 FPS on Neu3D). If these numbers are taken from the original papers, which used different GPUs and often different render resolutions, the reported 300–800% speedup is not verifiable. Please re-benchmark all methods on the same GPU, the same resolution, and the same codebase/measurement protocol, or explicitly state the source and conditions of each FPS number and narrow the speed claim accordingly.
  2. [Abstract and Section 4.2, Tables 1 and 2] The claim of achieving state-of-the-art storage efficiency 'without compromising rendering quality' is contradicted by the LPIPS results in the paper's own tables. On HyperNeRF, the three ADC-GS operating points have LPIPS 0.315, 0.278, and 0.252, versus 0.231 for E-D3DGS; on Neu3D, they are 0.066, 0.065, and 0.061, versus 0.030. This is a consistent and material perceptual-quality gap. Please revise the abstract and conclusions to say that PSNR and SSIM are comparable at reduced storage, and explicitly discuss the LPIPS trade-off.
  3. [Section 4.2 and contribution list] The statement 'up to 200× storage reduction over existing 4DGS methods' is not tied to a specific baseline in the text. Against the deformation-based baselines in Table 1, the reduction is approximately 11–16× (e.g., 47 MB for E-D3DGS versus 4.02 MB for ours). Against Real-Time4DGS in Table 2, the reduction is about 250×. Please specify which baseline yields the 200× figure and report per-baseline size ratios, so the claim is unambiguous and reproducible.
minor comments (6)
  1. [Section 3.4, Eq. (9)] The rate expression uses the notation 'MY' instead of a summation over the M chunks; please fix the typography and define all variables (e.g., the meaning of the product/sum over chunks).
  2. [Section 3.5, Eq. (11)] The temporal significance Ψ(k,t) is introduced in the main text, but its precise definition (the alpha-blending weight from Eq. (12) in Appendix A) appears only later. Please state the definition when the formula is first used.
  3. [Section 4.2, Table 6 and Table 7] The per-scene tables report PSNR and size but not SSIM or LPIPS; consider adding these metrics so that the per-scene behavior of perceptual quality can be assessed.
  4. [Appendix C] The coding time is reported as '1.27 seconds and 0.83 seconds under λe = e−2' without specifying the scene, resolution, or bitrate; please provide the experimental context.
  5. [Tables 4 and 5] The heading 'Ablation studies on the proportion of Gaussian primitives' is misleading; the parameter K is the number of primitives per anchor, not a proportion. Please rephrase.
  6. [Section 3.4] The text says 'λssim donates weighting coefficients'; 'donates' should be 'denotes'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the system is evaluated against external benchmarks and no derivation reduces to its own inputs.

full rationale

ADC-GS is an empirical systems paper. Its contributions (anchor-driven canonical representation, coarse-to-fine deformation, multi-dimension entropy model, temporal-significance anchor refinement) are defined by Eqs. (1)-(5), (6)-(8), (9)-(10), and (11)-(12). None of these equations is fitted to the headline metrics and then reported as a prediction: the anchor/residual prediction is a representation choice; the deformation networks are trained with reconstruction losses; the MEM estimates bitrate during training, but the reported sizes are the actual compressed bitstreams encoded with AE/G-PCC; FPS and PSNR are measured on HyperNeRF and Neu3D against external baselines. The only same-author citations are [Huang et al. 2024] for the alpha-blending weight in Eq. (11) and for a 3DGS compression technique in related work; these are reusable formulas/heuristics, not load-bearing uniqueness theorems or fitted inputs. The unverified part of the paper is the cross-hardware FPS comparison (baseline FPS values lack stated measurement conditions), which is an experimental-control issue rather than circular reasoning. Hence score 0.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central performance claim rests on standard volume-splatting rendering, COLMAP initialization, G-PCC compression, and the assumption that K=10 Gaussian primitives per anchor can locally represent the dynamic scene. The temporal significance heuristic comes from the authors' prior work (Huang et al. 2024). No free constants are hidden in a derivation; hyperparameters are standard ML choices and are disclosed.

free parameters (5)
  • K (Gaussians per anchor) = 10
    Chosen via ablation on HyperNeRF and Neu3D (Tables 4-5); K=10 balances PSNR vs size.
  • λe (rate-distortion Lagrange multiplier) = e^-2 to e^-4 (0.135 to 0.018)
    Varies the bitrate/quality trade-off, reported at three rates.
  • λssim = 0.2
    Fixed weighting for SSIM loss.
  • Quantization step sizes Qi = 0.1, 0.1, 0.01, 0.01 for fv, fg, Σv, Cv
    Hand-set per feature type to adjust quantization scale.
  • Feature dimensions Nv, Ng, M = 32, 16, 4
    Architecture choices; residual feature bitrate dominates, so chunking M=4 affects compression efficiency.
assumptions (5)
  • standard math Volume splatting as defined in 3DGS (Kerbl et al. 2023) accurately models radiance in dynamic scenes.
    The rendering equation from Kerbl et al. is used without modification; it is a standard background tool in the field.
  • domain assumption COLMAP sparse point clouds provide a sufficient anchor initialization for scenes tested.
    Anchors are initialized by downsampling COLMAP points; the method depends on this initialization being adequate for the datasets used.
  • domain assumption K=10 neural Gaussians per anchor can faithfully represent the local appearance and motion.
    The representation capacity of the scene is tied to this hyperparameter; the ablation supports K=10 for these datasets, but it is not proven general.
  • domain assumption Temporal significance weighting Ψ(k,t) from Huang et al. 2024 is a valid importance measure for growing/pruning anchors.
    The anchor refinement uses this heuristic to decide where to add or remove anchors; its validity is inherited from earlier work.
  • domain assumption G-PCC losslessly compresses anchor positions without affecting rendering.
    Anchor positions Xv are compressed with G-PCC and then fixed; the paper assumes this does not introduce distortion that would change quality.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction." pith.science (2026). https://pith.science/paper/WYQ2VLJA

@misc{pith2026250508196,
  author       = {Pith},
  title        = {Pith review of: ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WYQ2VLJA}},
  note         = {Machine review of arXiv:2505.08196}
}
read the original abstract

Existing 4D Gaussian Splatting methods rely on per-Gaussian deformation from a canonical space to target frames, which overlooks redundancy among adjacent Gaussian primitives and results in suboptimal performance. To address this limitation, we propose Anchor-Driven Deformable and Compressed Gaussian Splatting (ADC-GS), a compact and efficient representation for dynamic scene reconstruction. Specifically, ADC-GS organizes Gaussian primitives into an anchor-based structure within the canonical space, enhanced by a temporal significance-based anchor refinement strategy. To reduce deformation redundancy, ADC-GS introduces a hierarchical coarse-to-fine pipeline that captures motions at varying granularities. Moreover, a rate-distortion optimization is adopted to achieve an optimal balance between bitrate consumption and representation fidelity. Experimental results demonstrate that ADC-GS outperforms the per-Gaussian deformation approaches in rendering speed by 300%-800% while achieving state-of-the-art storage efficiency without compromising rendering quality. The code is released at https://github.com/H-Huang774/ADC-GS.git.

Figures

Figures reproduced from arXiv: 2505.08196 by the authors.

Figure 1
Figure 1. Comparison with concurrent dynamic scene reconstruc [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of local similarities of different features in [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of our ADC-GS framework. Top: ADC-GS organizes Gaussian primitives into a sparse set of anchors and compact residuals within canonical space. Bottom right: Gaussian primitives used for rendering are deformed from canonical space through a coarse￾to-fine strategy based on anchors. Bottom left: Rendering distortion and estimated bitrates from the MEM are jointly minimized to balance rendering quality and stor… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Illustration of the proposed MEM for accurate bitrates [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Rate-distortion curves of ADC-GS and comparison meth [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Bitstream analysis at multiple bitrate points on Neu3D. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Qualitative quality comparisons of “Flame Steak” (Top) in Neu3D and “Chicken” (Bottom) in HyperNeRF dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Ablation study of visual results for the coarse-to-fine de [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Detailed training process of our ADC-GS model. [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: Rate-distortion curves of ADC-GS and comparison meth [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: Qualitative results of HyperNeRF dataset. [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: Qualitative results of Neu3D dataset [PITH_FULL_IMAGE:figures/full_fig_p013_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 30 canonical work pages

  1. [1]

    Hyperreel: High-fidelity 6-dof video with ray-conditioned sampling

    Benjamin Attal, Jia-Bin Huang, Christian Richardt, Michael Zollhoefer, Johannes Kopf, Matthew O’Toole, and Changil Kim. Hyperreel: High-fidelity 6-dof video with ray-conditioned sampling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 16610--16620, 2023

    work page 2023

  2. [2]

    Per-gaussian embedding-based deformation for deformable 3d gaussian splatting

    Jeongmin Bae, Seoha Kim, Youngsik Yun, Hahyun Lee, Gun Bang, and Youngjung Uh. Per-gaussian embedding-based deformation for deformable 3d gaussian splatting. In European Conference on Computer Vision , pages 321--335. Springer, 2025

    work page 2025

  3. [3]

    Variational image compression with a scale hyperprior

    Johannes Ball \'e , David Minnen, Saurabh Singh, Sung Jin Hwang, and Nick Johnston. Variational image compression with a scale hyperprior. arXiv preprint arXiv:1802.01436 , 2018

    arXiv 2018

  4. [4]

    Hexplane: A fast representation for dynamic scenes

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

    work page 2023

  5. [5]

    MECD : Unlocking multi-event causal discovery in video reasoning

    Tieyuan Chen, Huabin Liu, Tianyao He, Yihang Chen, Chaofan Gan, Xiao Ma, Cheng Zhong, Yang Zhang, Yingxue Wang, Hui Lin, et al. MECD : Unlocking multi-event causal discovery in video reasoning. In The Thirty-eighth Annual Conference on Neural Information Processing Systems , 2024

    work page 2024

  6. [6]

    Fast feedforward 3d gaussian splatting compression

    Yihang Chen, Qianyi Wu, Mengyao Li, Weiyao Lin, Mehrtash Harandi, and Jianfei Cai. Fast feedforward 3d gaussian splatting compression. arXiv preprint arXiv:2410.08017 , 2024

    arXiv 2024

  7. [7]

    Hac: Hash-grid assisted context for 3d gaussian splatting compression

    Yihang Chen, Qianyi Wu, Weiyao Lin, Mehrtash Harandi, and Jianfei Cai. Hac: Hash-grid assisted context for 3d gaussian splatting compression. In European Conference on Computer Vision , pages 422--438. Springer, 2025

    work page 2025

  8. [8]

    D-dpcc: Deep dynamic point cloud compression via 3d motion prediction

    Tingyu Fan, Linyao Gao, Yiling Xu, Zhu Li, and Dong Wang. D-dpcc: Deep dynamic point cloud compression via 3d motion prediction. In Lud De Raedt, editor, Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22 , pages 898--904. International Joint Conferences on Artificial Intelligence Organization, 7 2022. Main Track

    work page 2022

Show all 42 references
  1. [9]

    Lightgaussian: Unbounded 3d gaussian compression with 15x reduction and 200+ fps, 2023

    Zhiwen Fan, Kevin Wang, Kairun Wen, Zehao Zhu, Dejia Xu, and Zhangyang Wang. Lightgaussian: Unbounded 3d gaussian compression with 15x reduction and 200+ fps, 2023

    2023

  2. [10]

    Fast dynamic radiance fields with time-aware neural voxels

    Jiemin Fang, Taoran Yi, Xinggang Wang, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Matthias Nie ner, and Qi Tian. Fast dynamic radiance fields with time-aware neural voxels. In SIGGRAPH Asia 2022 Conference Papers , pages 1--9, 2022

    2022

  3. [11]

    A hierarchical compression technique for 3d gaussian splatting compression

    He Huang, Wenjie Huang, Qi Yang, Yiling Xu, et al. A hierarchical compression technique for 3d gaussian splatting compression. arXiv preprint arXiv:2411.06976 , 2024

    2024 arXiv

  4. [12]

    Grid4D : 4D decomposed hash encoding for high-fidelity dynamic scene rendering

    Xu Jiawei, Fan Zexin, Yang Jian, and Xie Jin. Grid4D : 4D decomposed hash encoding for high-fidelity dynamic scene rendering. The Thirty-eighth Annual Conference on Neural Information Processing Systems , 2024

    2024

  5. [13]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk \"u hler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph. , 42(4):139--1, 2023

    2023

  6. [14]

    Compact 3d gaussian representation for radiance field

    Joo Chan Lee, Daniel Rho, Xiangyu Sun, Jong Hwan Ko, and Eunbyung Park. Compact 3d gaussian representation for radiance field. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 21719--21728, 2024

    2024

  7. [15]

    Neural 3d video synthesis from multi-view video

    Tianye Li, Mira Slavcheva, Michael Zollhoefer, Simon Green, Christoph Lassner, Changil Kim, Tanner Schmidt, Steven Lovegrove, Michael Goesele, Richard Newcombe, et al. Neural 3d video synthesis from multi-view video. In Proceedings of the IEEE/CVF Conference on Computer Vision...

    2022

  8. [16]

    Spacetime gaussian feature splatting for real-time dynamic view synthesis

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

    2024

  9. [17]

    Encoding auxiliary information to restore compressed point cloud geometry

    Gexin Liu, Jiahao Zhu, Dandan Ding, and Zhan Ma. Encoding auxiliary information to restore compressed point cloud geometry. In Kate Larson, editor, Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24 , pages 2189--2197. Internati...

    2024

  10. [18]

    Compgs: Efficient 3d scene representation via compressed gaussian splatting

    Xiangrui Liu, Xinju Wu, Pingping Zhang, Shiqi Wang, Zhu Li, and Sam Kwong. Compgs: Efficient 3d scene representation via compressed gaussian splatting. In Proceedings of the 32nd ACM International Conference on Multimedia , pages 2936--2944, 2024

    2024

  11. [19]

    Dn-4dgs: Denoised deformable network with temporal-spatial aggregation for dynamic scene rendering

    Jiahao Lu, Jiacheng Deng, Ruijie Zhu, Yanzhe Liang, Wenfei Yang, Tianzhu Zhang, and Xu Zhou. Dn-4dgs: Denoised deformable network with temporal-spatial aggregation for dynamic scene rendering. arXiv preprint arXiv:2410.13607 , 2024

    arXiv

  12. [20]

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

    Tao Lu, Mulin Yu, Linning Xu, Yuanbo Xiangli, Limin Wang, Dahua Lin, and Bo Dai. Scaffold-gs: Structured 3d gaussians for view-adaptive rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 20654--20664, 2024

    2024

  13. [21]

    Fastscene: Text-driven fast indoor 3d scene generation via panoramic gaussian splatting

    Yikun Ma, Dandan Zhan, and Zhi Jin. Fastscene: Text-driven fast indoor 3d scene generation via panoramic gaussian splatting. In Kate Larson, editor, Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24 , pages 1173--1181. Internat...

    2024

  14. [22]

    Nerf: Representing scenes as neural radiance fields for view synthesis

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM , 65(1):99--106, 2021

    2021

  15. [23]

    Compressed 3d gaussian splatting for accelerated novel view synthesis

    Simon Niedermayr, Josef Stumpfegger, and R \"u diger Westermann. Compressed 3d gaussian splatting for accelerated novel view synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 10349--10358, 2024

    2024

  16. [24]

    Nerfies: Deformable neural radiance fields

    Keunhong Park, Utkarsh Sinha, Jonathan T Barron, Sofien Bouaziz, Dan B Goldman, Steven M Seitz, and Ricardo Martin-Brualla. Nerfies: Deformable neural radiance fields. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 5865--5874, 2021

    2021

  17. [25]

    Hypernerf: A higher-dimensional representation for topologically varying neural radiance fields

    Keunhong Park, Utkarsh Sinha, Peter Hedman, Jonathan T Barron, Sofien Bouaziz, Dan B Goldman, Ricardo Martin-Brualla, and Steven M Seitz. Hypernerf: A higher-dimensional representation for topologically varying neural radiance fields. arXiv preprint arXiv:2106.13228 , 2021

    2021 arXiv

  18. [26]

    D-nerf: Neural radiance fields for dynamic scenes

    Albert Pumarola, Enric Corona, Gerard Pons-Moll, and Francesc Moreno-Noguer. D-nerf: Neural radiance fields for dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 10318--10327, 2021

    2021

  19. [27]

    Seitz, and Richard Szeliski

    Noah Snavely, Steven M. Seitz, and Richard Szeliski. Photo tourism: exploring photo collections in 3d. In ACM SIGGRAPH 2006 Papers , SIGGRAPH '06, page 835–846, New York, NY, USA, 2006. Association for Computing Machinery

    2006

  20. [28]

    Nerfplayer: A streamable dynamic scene representation with decomposed neural radiance fields

    Liangchen Song, Anpei Chen, Zhong Li, Zhang Chen, Lele Chen, Junsong Yuan, Yi Xu, and Andreas Geiger. Nerfplayer: A streamable dynamic scene representation with decomposed neural radiance fields. IEEE Transactions on Visualization and Computer Graphics , 29(5):2732--2742, 2023

    2023

  21. [29]

    3dgstream: On-the-fly training of 3d gaussians for efficient streaming of photo-realistic free-viewpoint videos

    Jiakai Sun, Han Jiao, Guangyuan Li, Zhanjie Zhang, Lei Zhao, and Wei Xing. 3dgstream: On-the-fly training of 3d gaussians for efficient streaming of photo-realistic free-viewpoint videos. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pa...

    2024

  22. [30]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems , 2017

    2017

  23. [31]

    Bovik, H.R

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

    2004

  24. [32]

    Mixed neural voxels for fast multi-view video synthesis

    Feng Wang, Sinan Tan, Xinghang Li, Zeyue Tian, Yafei Song, and Huaping Liu. Mixed neural voxels for fast multi-view video synthesis. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 19706--19716, 2023

    2023

  25. [33]

    Arithmetic coding for data compression

    Ian H Witten, Radford M Neal, and John G Cleary. Arithmetic coding for data compression. Communications of the ACM , 30(6):520--540, 1987

    1987

  26. [34]

    4d gaussian splatting for real-time dynamic scene rendering

    Guanjun Wu, Taoran Yi, Jiemin Fang, Lingxi Xie, Xiaopeng Zhang, Wei Wei, Wenyu Liu, Qi Tian, and Xinggang Wang. 4d gaussian splatting for real-time dynamic scene rendering. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 20310--20320, 2024

    2024

  27. [35]

    4d gaussian splatting with scale-aware residual field and adaptive optimization for real-time rendering of temporally complex dynamic scenes

    Jinbo Yan, Rui Peng, Luyang Tang, and Ronggang Wang. 4d gaussian splatting with scale-aware residual field and adaptive optimization for real-time rendering of temporally complex dynamic scenes. In Proceedings of the 32nd ACM International Conference on Multimedia , pages 7871...

    2024

  28. [36]

    A benchmark for gaussian splatting compression and quality assessment study

    Qi Yang, Kaifa Yang, Yuke Xing, Yiling Xu, and Zhu Li. A benchmark for gaussian splatting compression and quality assessment study. In Proceedings of the 6th ACM International Conference on Multimedia in Asia , pages 1--8, 2024

    2024

  29. [37]

    Real-time photorealistic dynamic scene representation and rendering with 4d gaussian splatting

    Zeyu Yang, Hongye Yang, Zijie Pan, and Li Zhang. Real-time photorealistic dynamic scene representation and rendering with 4d gaussian splatting. In International Conference on Learning Representations (ICLR) , 2024

    2024

  30. [38]

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

    Ziyi Yang, Xinyu Gao, Wen Zhou, Shaohui Jiao, Yuqing Zhang, and Xiaogang Jin. Deformable 3d gaussians for high-fidelity monocular dynamic scene reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 20331--20341, 2024

    2024

  31. [39]

    Efros, Eli Shechtman, and Oliver Wang

    Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , pages 586--595, 2018

    2018

  32. [40]

    Mega: Memory-efficient 4d gaussian splatting for dynamic scenes

    Xinjie Zhang, Zhening Liu, Yifan Zhang, Xingtong Ge, Dailan He, Tongda Xu, Yan Wang, Zehong Lin, Shuicheng Yan, and Jun Zhang. Mega: Memory-efficient 4d gaussian splatting for dynamic scenes. arXiv preprint arXiv:2410.13613 , 2024

    arXiv

  33. [41]

    Gaussianprediction: Dynamic 3d gaussian prediction for motion extrapolation and free view synthesis

    Boming Zhao, Yuan Li, Ziyu Sun, Lin Zeng, Yujun Shen, Rui Ma, Yinda Zhang, Hujun Bao, and Zhaopeng Cui. Gaussianprediction: Dynamic 3d gaussian prediction for motion extrapolation and free view synthesis. In ACM SIGGRAPH 2024 Conference Papers , pages 1--12, 2024

    2024

  34. [42]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence '...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.