Pith. sign in

REVIEW 6 major objections 5 minor 3 cited by

DynaSplat: Dynamic-Static Gaussian Splatting with Hierarchical Motion Decomposition for Scene Reconstruction

T0 review · 6 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read DynaSplat claims dynamic scene reconstruction is best done by first splitting the scene's Gaussians into static and dynamic sets using deformation statistics and camera-subtracted motion flow, then modeling the moving set with a…

desk verdict Plausible dynamic-GS pipeline with a real evaluation problem: the SOTA claim is contradicted by its own N3DV PSNR table, and the baseline rows are duplicated across datasets; still worth sending to reviewers if the authors clean it up. read the letter →

arxiv 2506.09836 v1 pith:C42CAAK4 submitted 2025-06-11 cs.CV cs.AI

Junli Deng , Ping Shi , Qipei Li , Jinyang Guo This is my paper
classification cs.CVcs.AI
keywords dynamicscenereconstruction3DGaussiansplattingdynamic-staticseparationhierarchicalmotiondecompositionopticalflowconsistencyphysically-basedopacityheadavatarnovelviewsynthesis
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 scene reconstruction asks for an accurate, temporally coherent 3D model built from video of moving content, and DynaSplat's answer is that the standard deformable-Gaussian pipeline wastes capacity by deforming every point even when most of a scene is still. The paper's claim is that a two-track design fixes this: classify each Gaussian as static or dynamic early in training, keep a single lightweight network for the static majority, and spend the expressive motion model only on the moving minority, whose motion is further decomposed into one coarse global component and one fine per-Gaussian residual. On the D-NeRF and N3DV benchmarks the method reports the best or near-best PSNR, SSIM, and LPIPS among the compared methods, and on a head-avatar benchmark it reports higher fidelity than specialized avatar baselines while training in about 11 minutes and rendering at 324 frames per second. If the claim holds, complex dynamic scenes and high-fidelity digital avatars can be reconstructed from video at interactive speeds without giving up accuracy.

What carries the argument

The load-bearing mechanism is the classification rule that produces the partition $G = G_{\text{dynamic}} \cup G_{\text{static}}$: a Gaussian is dynamic if its offset variance $\operatorname{Var}(\Delta\mu_g) > \tau$ and if, for more than a fraction $\gamma = 0.5$ of frames, its projected position falls on a pixel whose motion-flow magnitude exceeds $\epsilon = 1.0$. On top of that partition sits the hierarchical motion identity $\Delta X_g = \Delta X^{\text{coarse}}_g + \Delta X^{\text{fine}}_g$, where the coarse term averages motion over neighboring Gaussians and the fine term is an MLP applied to the coarse estimate plus a learned per-point feature, with four motion modes blended by learnable weights $\{\beta_m\}$. The third mechanism is physically-based opacity $\alpha_g = \alpha_0 \cos\theta_g / \|P_g - P_{\text{camera}}\|_2^2$ in place of the opacity reset used in vanilla 3D Gaussian Splatting, paired with a temporal importance filter that prunes Gaussians whose contribution to rendered pixels is negligible at every view and time. Together these carry the argument: the split focuses capacity on motion, the additive decomposition captures global trends plus local detail, and the opacity law handles occlusion, depth ordering, and semi-transparent surfaces.

What would settle it

On a sequence with a strongly moving camera whose pan direction parallels an independently moving object, residual camera-flow errors would land on static Gaussians: if DynaSplat flags a large share of the background as dynamic, its score should fall toward its own no-separation ablation (33.12 PSNR on D-NeRF). A second check is to sweep the separation threshold $\tau$ from $10^{-4}$ to $10^{-1}$ on D-NeRF and see whether the reported 34.39 PSNR and the margin over the closest competitor (34.25) survive the sweep.

Watch

Extended reading notes

Core claim

The central claim is that dynamic-static separation is the missing ingredient in deformable Gaussian splatting, where a scene is represented as thousands of small 3D ellipsoidal Gaussians, each with a position, rotation, scale, color, and opacity, projected and blended to render each frame. DynaSplat first trains a canonical-and-deformation model for a few thousand iterations, then labels a Gaussian as dynamic only when two independent signals agree: the temporal variance of its learned position offset exceeds a threshold $\tau$, and its projected positions keep landing on pixels where a motion flow map — optical flow with the camera-induced component removed — shows real movement. Static Gaussians retain a lightweight three-layer MLP; dynamic Gaussians are handed to a hierarchical model whose final motion is the sum of a coarse neighborhood-averaged offset and a fine MLP residual computed from that coarse estimate plus a learnable per-Gaussian feature, with several motion modes blended by learned weights. Opacity is no longer a free parameter: it is computed as $\alpha_g = \alpha_0 \cos\theta_g / \|P_g - P_{\text{camera}}\|_2^2$, and pruning is done by temporal importance across training views rather than by spatial attributes. With these pieces the paper reports PSNR 34.39 on D-NeRF against 34.25 for the closest Gaussian competitor, SSIM 0.967 and LPIPS 0.069 on N3DV, and PSNR 33.24 with 324 FPS inference on the head-avatar benchmark, and the ablation tables attribute the totals to all three components.

Load-bearing premise

The two-track model is only as good as the dynamic-static split, and that split inherits every error in the optical flow and the camera-motion subtraction, with hand-set thresholds ($\tau = 0.01$, $\epsilon = 1.0$, $\gamma = 0.5$) for which no sensitivity analysis is reported.

Editorial extensions

If this is right

  • Dynamic scenes can be rendered at interactive rates with the accuracy of slower methods: the head-avatar experiment reports 324 FPS inference after about 11 minutes of training, with PSNR 33.24 and LPIPS 0.0281 against specialized baselines.
  • Model capacity is spent where motion lives: static regions keep a single lightweight MLP while confirmed moving Gaussians alone receive the multi-mode hierarchical network, so background fidelity does not consume deformation parameters.
  • Motion that mixes a global body transform with local deformations — head pose plus facial expression, for instance — is exactly where the coarse-plus-fine sum pays off, which the paper argues is why the method suits avatar reconstruction.
  • Removing the physically-based opacity from the N3DV ablation alone drops SSIM from 0.967 to 0.930 and LPIPS from 0.069 to 0.150, indicating geometry-aware opacity and temporal importance pruning carry much of the perceptual gain on real-world video.
  • The D-NeRF table places the method ahead of non-Gaussian dynamic-scene methods too: the best non-Gaussian scores are K-Planes at 31.05 and HyperReel at 30.36, versus the proposed 34.39.

Reading between the lines

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

  • The most robust evidence for the headline claim may be the perceptual and structural gains on N3DV (LPIPS 0.069 and SSIM 0.967 versus 0.142 and 0.935 for the closest Gaussian competitor) rather than the 0.14 dB PSNR lead on D-NeRF, which is within the usual run-to-run spread of these pipelines.
  • Because the separation rule is a hand-set classifier over flow and deformation statistics, the fixed thresholds are the part of the design most likely to need per-scene tuning; replacing them with a learned separator is a natural and testable extension of the paper's own thesis.
  • The camera-subtracted flow map is the only motion cue for the dynamic-static split, so the method's real-world robustness is untested in exactly the regimes where optical flow struggles — heavy occlusion, textureless surfaces, and reflections — and those cases will decide how general the claim is.
  • The opacity formula is a simple physical prior that does not depend on motion at all; applying the same distance-and-angle opacity inside static Gaussian splatting would be a cheap way to test whether that contribution transfers beyond dynamic scenes.
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

6 major / 5 minor

Summary. The paper proposes DynaSplat, a dynamic 3D Gaussian Splatting method that combines three components: a dynamic-static separation mechanism based on per-Gaussian deformation-offset variance and 2D optical-flow consistency, a hierarchical motion model that blends multiple MLP motion modes with a coarse-to-fine decomposition, and a distance- and angle-dependent opacity computation. The method is evaluated on D-NeRF, N3DV, and a head-avatar benchmark, with quantitative comparisons to prior methods and ablations. The central claim is that DynaSplat surpasses state-of-the-art alternatives in accuracy and realism for dynamic scene reconstruction while remaining efficient.

Significance. If supported, the work would be a useful contribution to dynamic scene reconstruction: it extends 3DGS with a principled dynamic-static separation, a hierarchical motion representation, and a view-dependent opacity model, and it demonstrates competitive quantitative results on two standard benchmarks plus a downstream avatar application. The paper's strengths include a clear architectural description, ablations for each proposed component, and results on multiple datasets. However, the significance is currently limited by evaluation issues: an abstract-level accuracy claim that is contradicted by one of the paper's own tables, duplicated baseline rows across two different datasets, and an ablation row that shows almost no effect for one claimed core component. These issues need to be addressed before the significance claim can be accepted.

major comments (6)
  1. [Abstract and Table II] The abstract claims DynaSplat 'surpasses state-of-the-art alternatives in accuracy and realism,' but in Table II on the N3DV dataset the proposed method achieves PSNR 31.68, which is lower than the 4D-Rotor-Gaussians PSNR of 31.80 (though SSIM and LPIPS are better). The claim is therefore not supported on the PSNR axis for this dataset. The text should qualify the accuracy claim by metric or remove the blanket 'surpasses' statement.
  2. [Tables I and II] Tables I and II contain identical rows for DyNeRF (29.58/-/0.080), StreamRF (28.16/0.850/0.310), HyperReel (30.36/0.920/0.170), and NeRFPlayer (30.69/-/0.110) on two different datasets (D-NeRF synthetic, N3DV real). Identical metric values on different datasets are implausible and suggest the numbers were copied without re-evaluation or proper sourcing. The authors must verify the correct numbers for each dataset, state explicitly whether baselines were retrained or numbers were taken from prior papers, and correct any errors.
  3. [Section III-D, Eq. (14)] The 'physically-based opacity computation' is stated as alpha_g = alpha0 * cos(theta_g) / ||P_g - P_camera||^2 without derivation or implementation details. In particular, the paper does not explain how the Gaussian normal n_g is computed, how the opacity is normalized to valid [0,1] ranges, or why opacity should scale inversely with squared camera distance in a physically based volumetric rendering model. This is a load-bearing component of the method and the ablation table, so the authors should either provide a derivation and implementation specification or rename it as a heuristic.
  4. [Section IV-D, Table IV] The ablation for 'Without Physically-Based Opacity Computation' on the D-NeRF dataset reports SSIM 0.965 (identical to the full model) and LPIPS 0.035 (0.001 worse than the full model), while the text in Section IV-D-3 claims the opacity computation 'significantly improves' rendering. This discrepancy weakens the contribution claim. The authors should provide per-scene metric deltas and explain why the component has negligible effect on D-NeRF metrics while having a larger effect on N3DV, or re-scope the opacity contribution to specific visual effects.
  5. [Section III-B] Equations (11) and (12) are duplicates: both define G_dynamic with identical conditions, and Section III-B-c on obtaining the motion flow map is not integrated into the formal definition. This duplication is confusing and leaves unclear whether the motion flow map is part of the thresholding pipeline or only a verification step. The authors should merge the definitions and clarify the exact computational flow.
  6. [Section IV-C, Table III] The head-avatar experiment is presented as a significant application, but no dataset name, data preprocessing, training protocol, or evaluation split is provided for the 'benchmark dataset' in Table III. Without this information, the reported improvements over INSTA, FlashAvatar, and GaussianBS cannot be independently assessed. Please specify the dataset and protocol.
minor comments (5)
  1. [Section III-C, Eq. (13)] The blending parameters beta_m in the Adaptive Motion Networks are introduced but no constraint (e.g., softmax or sum-to-one) is specified; please clarify how the mixture is normalized.
  2. [Section III-D, Eq. (15)] The temporal importance w_i is defined as a maximum over all views and timestamps; please clarify whether this is recomputed during training or fixed, and how the pruning threshold tau=0.02 interacts with the removed opacity reset operation.
  3. [References] Reference [7] for 3D Gaussian Splatting lists an incorrect author sequence ('R. Kerbl, A. Ruiz, T. Leimkühler, C. Buehler, S. Saito, A. Vedaldi, ...'); the standard citation should be Kerbl, Kopanas, Leimkühler, and Drettakis. References [8] and [22] appear to describe the same work with different publication venues; please unify.
  4. [Throughout] There are several typographical issues: 'GroudTruth' in Figure 5, 'shown is Fig. 2' in Section III-D, and 'lightweight three MLP' in Section III-B should be 'three-layer MLP'. Please proofread.
  5. [Section II] The related work section is very brief and does not discuss several recent dynamic Gaussian Splatting methods beyond the ones directly compared; adding a short paragraph on these would help position the contributions.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: DynaSplat's headline claims are empirical comparisons on external benchmarks, and its components are heuristic designs rather than predictions derived from the target metrics.

full rationale

The paper's central claim is that DynaSplat surpasses state-of-the-art methods on PSNR/SSIM/LPIPS (Tables I-III). These metrics are external and are not used to define any network component. The dynamic-static separation (Sec. III-B) classifies Gaussians by thresholding learned deformation-offset variance and optical-flow magnitude; although the offset variance comes from the deformation network's own outputs, the classification is a heuristic that is then re-optimized, and it does not by construction determine any reported metric. The hierarchical motion decomposition (Sec. III-C) is a coarse-plus-fine MLP combination, and the physically-based opacity (Sec. III-D) is a fixed functional form with an importance-based pruning rule; neither equals the loss function or the evaluation metrics. The only self-referential aspect is that 'dynamic' is defined in terms of the learned deformation offsets, but that is self-consistency of a segmentation heuristic, not a derivation of the headline result from its inputs. Thresholds (tau=0.01, epsilon=1.0, gamma=0.5, pruning tau=0.02) are hand-set without sensitivity analysis, but that is an overfitting/robustness risk, not circularity. The ablation row 'Without Physically-Based Opacity Computation' shows a small change on D-NeRF (SSIM identical 0.965, LPIPS 0.035 vs 0.034), which weakens the opacity contribution claim but does not make the pipeline circular. Citations to MotionGS and 3DGS are external prior work, and no load-bearing self-citation or imported uniqueness theorem appears. The paper would benefit from per-scene results and error bars, but those are validation-quality concerns, not circularity.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

Free parameters are hand-set thresholds and constants that the central pipeline depends on; none are derived from first principles or tested by sensitivity analysis. Axioms are standard 3DGS background, the dynamic-static distribution hypothesis, reliable flow after camera removal, and the ad hoc opacity model.

free parameters (7)
  • tau_offset (offset variance threshold) = 0.01
    Hand-set threshold in Eq. (8) to flag candidate dynamic Gaussians; no sensitivity analysis is provided.
  • epsilon (motion flow magnitude threshold) = 1.0
    Threshold in Eq. (9) for pixels considered moving; chosen by hand.
  • gamma (dynamic confirmation ratio) = 0.5
    Minimum fraction of frames where flow magnitude exceeds epsilon; set in Section IV-A.
  • tau_prune (temporal importance pruning threshold) = 0.02
    Pruning threshold for wi in Eq. (15); hand-set after 20k iterations.
  • alpha0 (initial opacity constant) = 1.0
    Constant in Eq. (14); the physically-based opacity model is not derived.
  • lambda (loss balance) = 0.1
    Weight in Eq. (16) between L1 and SSIM terms.
  • M (number of motion modes) = 4
    Number of MLPs blended in Adaptive Motion Networks; no ablation on M is reported.
assumptions (5)
  • standard math 3D Gaussian splatting rendering uses alpha compositing (Eq. 2) with per-Gaussian color, opacity, and covariance.
    Used as the underlying renderer; taken from 3DGS.
  • domain assumption Dynamic and static scene elements follow distinct motion distributions.
    Core hypothesis stated in the introduction that motivates separating Gaussians.
  • domain assumption The 2D motion flow map, after removing camera-induced flow, reveals true object motion.
    Invoked in Section III-B-c to confirm candidate dynamic Gaussians; depends on off-the-shelf optical flow and known poses and depth.
  • ad hoc to paper Opacity is inversely proportional to squared camera distance and proportional to cos(theta) (Eq. 14).
    Presented as physically based but no derivation from radiometry or volume rendering is given; alpha0 is a fitted constant.
  • domain assumption Gaussian positions are updated by learned offsets per frame via a canonical plus deformation network (Eqs. 3 through 5).
    Standard deformable-GS setup inherited from prior work; assumed to be a valid motion parametrization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DynaSplat: Dynamic-Static Gaussian Splatting with Hierarchical Motion Decomposition for Scene Reconstruction." pith.science (2026). https://pith.science/paper/C42CAAK4

@misc{pith2026250609836,
  author       = {Pith},
  title        = {Pith review of: DynaSplat: Dynamic-Static Gaussian Splatting with Hierarchical Motion Decomposition for Scene Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C42CAAK4}},
  note         = {Machine review of arXiv:2506.09836}
}
read the original abstract

Reconstructing intricate, ever-changing environments remains a central ambition in computer vision, yet existing solutions often crumble before the complexity of real-world dynamics. We present DynaSplat, an approach that extends Gaussian Splatting to dynamic scenes by integrating dynamic-static separation and hierarchical motion modeling. First, we classify scene elements as static or dynamic through a novel fusion of deformation offset statistics and 2D motion flow consistency, refining our spatial representation to focus precisely where motion matters. We then introduce a hierarchical motion modeling strategy that captures both coarse global transformations and fine-grained local movements, enabling accurate handling of intricate, non-rigid motions. Finally, we integrate physically-based opacity estimation to ensure visually coherent reconstructions, even under challenging occlusions and perspective shifts. Extensive experiments on challenging datasets reveal that DynaSplat not only surpasses state-of-the-art alternatives in accuracy and realism but also provides a more intuitive, compact, and efficient route to dynamic scene reconstruction.

Figures

Figures reproduced from arXiv: 2506.09836 by the authors.

Figure 1
Figure 1. Overview of DynaSplat. Our method first performs dynamic-static separation through offset variance analysis and 2D [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Illustration of physically-based opacity computation. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 4
Figure 4. Comparison of scenes from N3DV dataset. The success is primarily attributed to our Hierarchical Mo￾tion Modeling. Human head dynamics are a perfect example of composite motion, which our coarse-to-fine decomposition is ideally suited to handle. • The coarse motion component effectively captures the large-scale, rigid transformations of the head, such as nodding, shaking, and tilting. By averaging motion in￾formation… view at source ↗
Figures from the paper (3 more)
Figure 3
Figure 3. Figure 3: Comparison of scenes from D-NeRF dataset. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]
Figure 6
Figure 6. Figure 6: Effect of Dynamic-Static Separation. GroundTruth w/o Hierarchical w/ Hierarchical [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: Effect of Hierarchical Motion Modeling. GroundTruth w/o PhysicalOpacity w/ PhysicalOpacity [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Graphical X Splatting (GraphiXS): A Graphical Model for 4D Gaussian Splatting under Uncertainty

    cs.GR 2026-01 unverdicted novelty 7.0 of 10

    GraphiXS is a new probabilistic graphical framework that augments 4D Gaussian Splatting to systematically handle multiple types of data uncertainty such as view sparsity and missing frames.

  2. Does it matter which Gaussians you pick in 4D Gaussian streaming?

    cs.CV 2026-03 conditional novelty 4.0 of 10

    A reinforcement-learned plug-in sampler can match or beat IGS@8192 quality on N3DV and MeetingRoom using as few as 256 anchors while reducing per-frame time.

  3. SplitGaussian: Reconstructing Dynamic Scenes via Visual Geometry Decomposition

    cs.CV 2025-08 unverdicted novelty 4.0 of 10

    SplitGaussian reconstructs dynamic 3D scenes from monocular video by decomposing Gaussians into a rigid static branch and a deformable dynamic branch, claiming better motion separation and rendering quality than prior...

Reference graph

Works this paper leans on

35 extracted references · 30 canonical work pages · cited by 3 Pith papers

  1. [1]

    Hartley and A

    R. Hartley and A. Z. E. Z, Multiple View Geometry in Computer Vision, 2nd ed. Cambridge University Press, 2004

    work page 2004

  2. [2]

    Simultaneous localization and map- ping (slam): Part i the essential algorithms,

    H. Durrant-Whyte and T. Bailey, “Simultaneous localization and map- ping (slam): Part i the essential algorithms,” IEEE Robotics & Automa- tion Magazine, vol. 13, no. 2, pp. 99–110, 2006. TABLE IV: Ablation Study Results on D-NeRF and N3DV Video Datasets Method D-NeRF Dataset N3DV Video Dataset PSNR ↑ SSIM ↑ LPIPS ↓ PSNR ↑ SSIM ↑ LPIPS ↓ Full Model 34.39 ...

    work page 2006

  3. [3]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” in ECCV, 2020, pp. 405–421

    work page 2020

  4. [4]

    Nerfies: Deformable neural radiance fields,

    J. J. Park and et al., “Nerfies: Deformable neural radiance fields,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 4480–4490

    work page 2021

  5. [5]

    Freenerf: Improving few-shot neural rendering with free frequency regularization,

    J. Yang, M. Pavone, and Y . Wang, “Freenerf: Improving few-shot neural rendering with free frequency regularization,” in CVPR, 2023

    work page 2023

  6. [6]

    Nerf dynamic scene reconstruction based on motion, semantic information and inpainting,

    J. Liu, H. Cheng, S. Wang, F. Zhao, and M. Li, “Nerf dynamic scene reconstruction based on motion, semantic information and inpainting,” Neurocomputing, vol. 630, p. 129653, 2025. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S092523122500325X

    work page 2025

  7. [7]

    3d gaussian splatting for real-time radiance field rendering,

    B. Kerbl, A. Ruiz, T. Leimkühler, C. Buehler, S. Saito, A. Vedaldi, T. Weyrich, H.-P. Seidel, B. Bickel, M. Rittig et al. , “3d gaussian splatting for real-time radiance field rendering,” ACM Trans. Graph. (SIGGRAPH), vol. 42, no. 4, 2023

    work page 2023

  8. [8]

    Motion-aware 3d gaussian splatting for efficient dynamic scene reconstruction,

    Z. Guo, W. Zhou, L. Li, M. Wang, and H. Li, “Motion-aware 3d gaussian splatting for efficient dynamic scene reconstruction,” IEEE Transactions on Circuits and Systems for Video Technology , pp. 1–1, 2024

    work page 2024

Show all 35 references
  1. [9]

    Motiongs: Exploring explicit motion guidance for deformable 3d gaussian splatting,

    R. Zhu, Y . Liang, H. Chang, J. Deng, J. Lu, W. Yang, T. Zhang, and Y . Zhang, “Motiongs: Exploring explicit motion guidance for deformable 3d gaussian splatting,” NeurIPS, 2024

    2024

  2. [10]

    Unsupervised 3d part decomposition via leveraged gaussian splatting,

    J. Choy, G. Cha, H. Kee, and S. Oh, “Unsupervised 3d part decomposition via leveraged gaussian splatting,” 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pp. 2647–2652, 2024. [Online]. Available: https://api.semanticscholar.org/ CorpusID:275021195

    2024

  3. [11]

    Photorealistic scene reconstruction by voxel coloring,

    S. M. Seitz and C. R. Dyer, “Photorealistic scene reconstruction by voxel coloring,” International journal of computer vision , vol. 35, pp. 151–173, 1999

    1999

  4. [12]

    Depth map prediction from a single image using a multi-scale deep network,

    D. Eigen et al. , “Depth map prediction from a single image using a multi-scale deep network,” in Proceedings of the Advances in Neural Information Processing Systems (NeurIPS) , 2014, pp. 2366–2374

    2014

  5. [13]

    D-nerf: Neural radiance fields for dynamic scenes,

    A. Pumarola, D. Azinovic, R. Martin-Brualla, G. Pons-Moll, and M. Zollhöfer, “D-nerf: Neural radiance fields for dynamic scenes,” in CVPR, 2021, pp. 10 318–10 327

    2021

  6. [14]

    Neural scene flow fields for space-time view synthesis of dynamic scenes,

    Z. Li, S. Niklaus, N. Snavely, and O. Wang, “Neural scene flow fields for space-time view synthesis of dynamic scenes,” in CVPR, 2021, pp. 6498–6508

    2021

  7. [15]

    K-planes: Explicit radiance fields in space, time, and appearance,

    S. Fridovich-Keil, G. Meanti, F. R. Warburg, B. Recht, and A. Kanazawa, “K-planes: Explicit radiance fields in space, time, and appearance,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 12 479–12 488

    2023

  8. [16]

    4d gaussian splatting for real-time dynamic scene rendering,

    G. Wu, T. Yi, J. Fang, L. Xie, X. Zhang, W. Wei, W. Liu, Q. Tian, and W. Xinggang, “4d gaussian splatting for real-time dynamic scene rendering,” in CVPR, 2024

    2024

  9. [17]

    Mixed 3d gaussian for dynamic scenes representation and rendering,

    J. Guo, T. Wang, and C. Wang, “Mixed 3d gaussian for dynamic scenes representation and rendering,” in 2024 IEEE International Conference on Multimedia and Expo (ICME) . IEEE, 2024, pp. 1–6

    2024

  10. [18]

    Swings: sliding windows for dy- namic 3d gaussian splatting,

    R. Shaw, M. Nazarczuk, J. Song, A. Moreau, S. Catley-Chandar, H. Dhamo, and E. Pérez-Pellitero, “Swings: sliding windows for dy- namic 3d gaussian splatting,” in ECCV, 2024

    2024

  11. [19]

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

    Z. Yang, X. Gao, W. Zhou, S. Jiao, Y . Zhang, and X. Jin, “Deformable 3d gaussians for high-fidelity monocular dynamic scene reconstruction,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 20 331–20 341

    2024

  12. [20]

    Event-boosted deformable 3d gaussians for fast dynamic scene reconstruction,

    W. Xu, W. Weng, Y . Zhang, R. Xu, and Z. Xiong, “Event-boosted deformable 3d gaussians for fast dynamic scene reconstruction,” arXiv preprint arXiv:2411.16180, 2024

    2024 arXiv

  13. [21]

    Memflow: Optical flow estimation and prediction with memory,

    Q. Dong and Y . Fu, “Memflow: Optical flow estimation and prediction with memory,” in Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , 2024

    2024

  14. [22]

    Motion-aware 3d gaus- sian splatting for efficient dynamic scene reconstruction,

    Z. Guo, W. Zhou, L. Li, M. Wang, and H. Li, “Motion-aware 3d gaus- sian splatting for efficient dynamic scene reconstruction,” in European Conference on Computer Vision , 2024

    2024

  15. [23]

    Gaussian-flow: 4d reconstruction with dynamic 3d gaussian particle,

    Y . Lin, Z. Dai, S. Zhu, and Y . Yao, “Gaussian-flow: 4d reconstruction with dynamic 3d gaussian particle,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 21 136–21 145

    2024

  16. [24]

    Superpoint gaussian splatting for real-time high-fidelity dynamic scene reconstruction,

    G. Z. Diwen Wan, Ruijie Lu, “Superpoint gaussian splatting for real-time high-fidelity dynamic scene reconstruction,” in Forty-first International Conference on Machine Learning , 2024

    2024

  17. [25]

    Neural 3d video synthesis from multi-view video,

    T. Li, M. Slavcheva, M. Zollhoefer, S. Green, C. Lassner, C. Kim, T. Schmidt, S. Lovegrove, M. Goesele, R. Newcombe et al., “Neural 3d video synthesis from multi-view video,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 5521–5531

    2022

  18. [26]

    The unreasonable effectiveness of deep features as a perceptual metric,

    R. Zhang, P. Isola, A. A. Efros, E. Shechtman, and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 586–595

    2018

  19. [27]

    Streaming radiance fields for 3d video synthesis,

    L. Li, Z. Shen, zhongshu wang, L. Shen, and P. Tan, “Streaming radiance fields for 3d video synthesis,” in Advances in Neural Information Processing Systems, A. H. Oh, A. Agarwal, D. Belgrave, and K. Cho, Eds., 2022

    2022

  20. [28]

    HyperReel: High-fidelity 6-DoF video with ray- conditioned sampling,

    B. Attal, J.-B. Huang, C. Richardt, M. Zollhoefer, J. Kopf, M. O’Toole, and C. Kim, “HyperReel: High-fidelity 6-DoF video with ray- conditioned sampling,” in Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    2023

  21. [29]

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

    L. Song, A. Chen, Z. Li, Z. Chen, L. Chen, J. Yuan, Y . Xu, and A. Geiger, “Nerfplayer: A streamable dynamic scene representation with decomposed neural radiance fields,” IEEE Transactions on Visualization and Computer Graphics , 2023

    2023

  22. [30]

    4d-rotor gaussian splatting: Towards efficient novel view synthesis for dynamic scenes,

    Y . Duan, F. Wei, Q. Dai, Y . He, W. Chen, and B. Chen, “4d-rotor gaussian splatting: Towards efficient novel view synthesis for dynamic scenes,” in Proc. SIGGRAPH, 2024

    2024

  23. [31]

    Mixed neural voxels for fast multi-view video synthesis,

    F. Wang, S. Tan, X. Li, Z. Tian, Y . Song, and H. Liu, “Mixed neural voxels for fast multi-view video synthesis,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , October 2023, pp. 19 706–19 716

    2023

  24. [32]

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

    Z. Yang, H. Yang, Z. Pan, and L. Zhang, “Real-time photorealistic dynamic scene representation and rendering with 4d gaussian splatting,” in The Twelfth International Conference on Learning Representations

  25. [33]

    Instant Volumetric Head Avatars, 2023

    2023

  26. [34]

    Flashavatar: High-fidelity head avatar with efficient gaussian embedding,

    J. Xiang, X. Gao, Y . Guo, and J. Zhang, “Flashavatar: High-fidelity head avatar with efficient gaussian embedding,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 2024

    2024

  27. [35]

    3d gaussian blendshapes for head avatar animation,

    S. Ma, Y . Weng, T. Shao, and K. Zhou, “3d gaussian blendshapes for head avatar animation,” in ACM SIGGRAPH Conference Proceedings, Denver, CO, United States, July 28 - August 1, 2024 , 2024

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.