Pith. sign in

REVIEW 5 major objections 4 minor 27 references

SHaDe: Compact and Consistent Dynamic 3D Reconstruction via Tri-Plane Deformation and Latent Diffusion

T0 review · 5 major / 4 minor · reviewed 2026-08-07 · 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 dynamic 3D reconstruction from sparse multi-view video can be made compact, explicit, and temporally consistent by replacing motion MLPs with a tri-plane offset field, decoding appearance with spherical-harmonics…

desk verdict A coherent three-module architecture with an honest related work section, but the paper's central claim—a fully explicit, MLP-free deformation field—is contradicted by its own inference path, and too many load-bearing details are unspecified to support the SOTA claim. read the letter →

arxiv 2505.16535 v1 pith:5AEW5P6H submitted 2025-05-22 cs.CV

Asrar Alruwayqi This is my paper · ORCID
classification cs.CV
keywords dynamic3Dreconstructiontri-planedeformationfieldsphericalharmonicsattentionlatentdiffusionsparse-viewtemporalconsistency4Dscenerepresentationvolumerendering
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

The paper attempts to show that dynamic 3D reconstruction from sparse multi-view video does not need learned deformation MLPs or large MLP radiance decoders. Its recipe is to store time-varying features on three orthogonal 2D planes, warp each query point into canonical space with a linear projection of summed plane features, and decode color with a spherical-harmonics attention head, while a latent diffusion module refines the tri-plane latents. If this is right, 4D reconstruction becomes more compact (27.1M parameters, 6.4 GB peak memory) and more robust to sparse inputs, beating HexPlane and 4D Gaussian Splatting on the D-NeRF benchmark in PSNR, SSIM, and LPIPS. The reason to care is that the same components could make high-quality dynamic reconstruction practical with only a handful of cameras.

What carries the argument

Three mechanisms carry the argument. The explicit tri-plane deformation field: three 256x256x32 time-conditioned planes $F_{xy}$, $F_{yz}$, $F_{xz}$; query features are interpolated, summed, and linearly projected by $\Delta x = W(f_{xy}+f_{yz}+f_{xz})+b$ and then $x_c = x + \Delta x$, with no MLP or nonlinearity in the motion path. The SH-attention radiance head: stored SH coefficients up to order $L=4$ are reweighted by learned view- and time-dependent attention weights $\alpha_{lm}(d,t)$ before composing color as $c(d,t)=\sum_{l,m} \alpha_{lm}(d,t)\, c_{lm}\, Y_{lm}(d)$, with density coming from a separate tri-plane. The latent diffusion refinement path: tri-plane patches become 768 tokens, a transformer pools them to a 512-dimensional latent, a time-conditioned 3D U-Net denoises that latent with a DDPM objective, a decoder returns refined features and offsets, and a temporal-consistency loss encourages smooth latent evolution.

What would settle it

Inspect the released code or weights for Eq. (2). If $W$ and $b$ change during training, the 'fixed, non-learned' claim is false; if they are frozen, then a controlled experiment that randomly re-initializes and freezes $W$ and $b$ and retrains the rest of the model on the T-Rex scene should reproduce the reported 35.8 dB PSNR, and if it does not, the deformation field is not a fixed parameter-free component as described.

Watch

Extended reading notes

Core claim

The paper's central discovery claim is that a 4D scene can be carried by three time-conditioned 2D feature planes whose summed interpolated features map, through the fixed linear projection of Eq. (2), to a per-point deformation offset, so all motion is an explicit warp into canonical space with no MLP in the motion path. In canonical space, a light SH-attention head computes color as a sum over SH bands with learned view- and time-dependent weights on stored coefficients, and a separate tri-plane gives density. A transformer pools plane patches into a 512-dimensional latent, a DDPM-style diffusion process trained with a DDIM sampler and a temporal-consistency loss refines that latent, and the refined features and offsets feed differentiable volume rendering. On four D-NeRF scenes the paper reports higher PSNR and SSIM and lower LPIPS than HexPlane and 4D-GS, with 27.1M parameters and 6.4 GB peak memory, and retains quality when input views drop to 3-5.

Load-bearing premise

The load-bearing premise is that the fixed linear projection in Eq. (2) can actually express a scene's motion; the paper never says how the matrix $W$ and bias $b$ are set, and if they are truly fixed constants the deformation field cannot adapt to different scenes, while if they are learned the 'fully explicit, non-learned' claim is false.

Editorial extensions

If this is right

  • Dynamic reconstruction becomes an explicit, inspectable warp: the motion path is a linear projection of plane features rather than a trained network, which cuts memory and inference cost.
  • Sparse capture becomes viable: the paper reports strong results from 3-5 views on D-NeRF, where the original D-NeRF method uses 100 or more views.
  • The SH-attention color head makes appearance interpretable: which SH frequency bands dominate a given view and time can be read directly from the attention weights.
  • The latent diffusion prior should stabilize fast or ambiguous motion, turning underdetermined tri-plane representations into coherent temporal sequences rather than letting the renderer hallucinate arbitrary geometry.

Reading between the lines

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

  • The 'no MLP' claim is scoped to the deformation path: the SH attention weights, the transformer, and the diffusion denoiser are themselves learned networks, so the whole system is not MLP-free.
  • Because $W$ and $b$ in Eq. (2) are never specified, a reader cannot verify that the deformation field is fixed; if these parameters are actually optimized, the central architectural claim is overstated.
  • The paper's gains are shown only on synthetic D-NeRF scenes, so the sparse-view robustness claim has not yet been tested on real-world video with occlusions; the modular design makes that extension a natural next experiment.
  • Varying the number of DDIM steps at test time should interpolate between fidelity to input views and smoothness of the learned prior, giving a tunable robustness knob the paper does not quantify.
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

5 major / 4 minor

Summary. The paper proposes SHaDe, a framework for dynamic 3D reconstruction that combines three components: an explicit tri-plane deformation field, a canonical radiance field with spherical-harmonics (SH) attention, and a latent diffusion refinement module. The method is evaluated on synthetic D-NeRF scenes against HexPlane and 4D Gaussian Splatting, with additional ablations on the T-Rex scene. The central claims are that deformation is computed by a fixed, non-learned linear projection of tri-plane features, that SH attention replaces MLP decoders, and that a transformer-guided latent diffusion prior improves robustness under sparse views and out-of-distribution motion. The paper reports state-of-the-art PSNR/SSIM/LPIPS numbers and efficiency measurements. As written, however, several load-bearing architectural statements are internally inconsistent or underspecified, particularly the relationship between the explicit deformation of Eq. (2) and the learned decoder output of Eq. (6), the status of W and b, and the sparse-view evaluation protocol.

Significance. If the claims were fully substantiated, the work would offer a useful direction: compact explicit deformation via tri-planes, an interpretable SH attention radiance head, and diffusion-based latent refinement are all relevant to the dynamic reconstruction community. The reported parameter and memory counts (Table 2) are concrete positives, and the modular ablation structure is sensible. However, the paper provides no code, no error bars, and no reproducible specification of the core projection in Eq. (2), and the final rendering path appears to use the learned decoder output of Sec. 3.3 rather than the explicit offset. These issues prevent verification of the state-of-the-art claim and of the claimed MLP-free deformation mechanism. The significance is therefore conditional on a major clarification and on additional experimental detail.

major comments (5)
  1. [Sec. 3.1, Eq. (2)] The deformation offset is defined as Delta x = W(fxy + fyz + fxz) + b with W in R^{3x32}, b in R^3, and the text says W and b are a 'fixed, non-learned linear projection.' The values or initialization of W and b are never given, so the model is not reproducible. Moreover, because the tri-plane features fxy, fyz, and fxz are learned, the resulting offset is a learned function of (x, t) even if W and b are fixed; the statement in Sec. 3.1 that 'the deformation is computed without any learnable components' is therefore misleading and should be corrected or justified.
  2. [Sec. 3.3, Eq. (6) and Fig. 2] The paper states at the end of Sec. 3.3 that 'the resulting refined features F_hat, Delta_hat are then used for differentiable volume rendering,' and Eq. (6) defines (F_hat, Delta_hat) = D(z) with a learned decoder D. The rendering path in Fig. 2, however, uses the original offset Delta x from Eq. (2). Unless D is the identity (which is not stated), the deformation actually used for rendering is the output of a learned network, directly contradicting the central claim in the Abstract and Sec. 3.1 that motion is modeled by a fully explicit, MLP-free deformation field. The paper must specify which offset is rendered, give the architecture and training details of D, and reconcile this with the explicit-deformation claim.
  3. [Sec. 3.2, Eq. (3)] The mapping from tri-plane features to the SH coefficients {c_lm} is not defined. The text says SH coefficients are 'stored in the tri-plane grid,' but it does not specify how they are retrieved, interpolated, or associated with canonical points, nor how the per-band attention weights alpha_lm(d, t) are parameterized beyond saying they come from a 'lightweight MLP.' This MLP also conflicts with the paper's repeated claim of avoiding MLPs; the MLP-free statement should be scoped to the deformation field only, and Eq. (3) should be made fully concrete.
  4. [Sec. 4, Fig. 3 and Tables 1-3] The sparse-view evaluation is not described as a protocol. The paper reports results for 3, 5, 10, and 20 views in Fig. 3 but does not state how those views were selected, whether HexPlane and 4D-GS were retrained under the same sparse-view conditions, or whether the D-NeRF baselines used their default dense-view training. Tables 1-3 also report no standard deviations or number of seeds, so the claimed superiority over HexPlane and 4D-GS is not statistically supported. These details are necessary to evaluate the central claim of robustness to sparse dynamic inputs.
  5. [Sec. 3.3 and Sec. 4] The latent diffusion prior is trained 'using only the same synthetic multi-view data (e.g., D-NeRF)' and evaluated on D-NeRF, so the claimed generalization to out-of-distribution motion is not demonstrated by any held-out OOD experiment. Furthermore, the inference procedure injects controlled Gaussian noise into the initial latent z0 and applies DDIM denoising, but the relationship between this noisy test-time latent and the training-time diffusion objective in Eq. (7) is never analyzed. The paper should either provide OOD experiments or remove the OOD robustness claim.
minor comments (4)
  1. [Related Work and Sec. 3.2] The method 'Planeoxel' in Sec. 2 should be 'Plenoxel' to match reference [25] and standard terminology.
  2. [Sec. 3.1] The phrase 'without any MLPs' in the module listing and 'without any learnable components' in Sec. 3.1 are inconsistent with the MLP for SH attention in Sec. 3.2 and with the learned tri-plane features; the claims should be reworded to state precisely which components contain no MLPs.
  3. [Sec. 4, Table 1] The table reports results on four scenes but does not say which scenes from D-NeRF are included or whether they are the standard test split; this should be stated for reproducibility.
  4. [Sec. 5, Table 3] The ablation 'w/o Deformation' removes the deformation field while retaining a dynamic model, but the paper does not describe how motion is represented in that variant; a one-sentence description of each ablation's architecture would improve interpretability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: benchmark comparisons are external and the core equations do not reduce to their inputs; the W/b omission and Eq. (2) vs Eq. (6) inconsistency are correctness issues, not circularity.

full rationale

The paper's derivation chain is not circular in the sense defined by the analysis protocol. Eq. (2) defines the deformation offset as a fixed linear projection of tri-plane features; Eq. (3) defines SH-attention color decoding; Eq. (7) is a standard DDPM objective; Eq. (9) combines losses. None of these equations are defined in terms of the reported benchmark numbers, and no fitted parameter is renamed as a prediction. The state-of-the-art claims are benchmarked against external methods (HexPlane, 4D-GS) on the D-NeRF dataset, which makes them falsifiable empirical claims rather than tautologies. Citations to EG3D, K-Planes, HexPlane, and other prior works are used only for standard components (tri-plane factorization, SH decoding, DDPM), and there is no self-citation chain or author-imported uniqueness theorem that forces the design choice. The paper does have serious non-circular weaknesses: W and b in Eq. (2) are never specified, so the 'fixed, non-learned' deformation is not fully defined; and the final rendering path uses refined offsets from the learned decoder D (Eq. (6), Inference section), which contradicts the claim that deformation is MLP-free and explicit. Additionally, the diffusion prior is trained on D-NeRF data and evaluated on D-NeRF, so the claimed out-of-distribution robustness is not demonstrated by an independent split. These are completeness and validity concerns, not definitional circularity, and therefore do not raise the circularity score.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method relies on a standard volume-rendering framework, but the tri-plane deformation and SH decoding introduce several hand-chosen constants and unresolved parameters. The most consequential assumptions are the representational capacity of linear tri-plane deformation (Sec. 3.1) and the validity of an in-distribution diffusion prior for OOD refinement (Sec. 3.3).

free parameters (5)
  • Plane resolution and channels = 256x256, 32 channels
    Chosen by hand in Sec. 3.1; representation capacity and memory depend on these values.
  • SH order L = L = 4
    Hand-chosen truncation in Eq. (3); affects expressiveness of view-dependent color.
  • Deformation projection W and bias b = Not specified
    Eq. (2) treats W and b as fixed but gives no values; if learned, they are fitted parameters.
  • Denoising steps T = 10
    Hand-chosen in Sec. 3.3 to keep the 20% runtime overhead.
  • Loss weights lambda_rec, lambda_diff, lambda_temporal = Not reported
    Eq. (9) combines three losses but relative weights are never given.
assumptions (4)
  • domain assumption A dynamic scene can be encoded by three time-modulated 2D feature planes and an explicit linear warp to canonical space (Eqs. 1-2).
    Sec. 3.1 adopts this decomposition from EG3D/K-Planes without proof that all target motions are representable by a linear projection of summed plane features.
  • domain assumption SH coefficients c_lm up to order 4 are stored on the tri-plane grid and interpolation yields valid coefficients (Eq. 3).
    Sec. 3.2 assumes this storage/decoding without specifying the mapping from 32-channel features to the 25 SH coefficients.
  • ad hoc to paper A DDPM/DDIM diffusion model trained on the same D-NeRF data provides a valid prior for test-time refinement.
    Sec. 3.3 trains the diffusion module on the same data, so OOD generalization is assumed rather than evidenced.
  • standard math Volume rendering with the predicted density and color produces photorealistic images under the benchmark's camera trajectories.
    Sec. 3 uses standard differentiable volume rendering from the NeRF literature without re-derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SHaDe: Compact and Consistent Dynamic 3D Reconstruction via Tri-Plane Deformation and Latent Diffusion." pith.science (2026). https://pith.science/paper/5AEW5P6H

@misc{pith2026250516535,
  author       = {Pith},
  title        = {Pith review of: SHaDe: Compact and Consistent Dynamic 3D Reconstruction via Tri-Plane Deformation and Latent Diffusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5AEW5P6H}},
  note         = {Machine review of arXiv:2505.16535}
}
read the original abstract

We present a novel framework for dynamic 3D scene reconstruction that integrates three key components: an explicit tri-plane deformation field, a view-conditioned canonical radiance field with spherical harmonics (SH) attention, and a temporally-aware latent diffusion prior. Our method encodes 4D scenes using three orthogonal 2D feature planes that evolve over time, enabling efficient and compact spatiotemporal representation. These features are explicitly warped into a canonical space via a deformation offset field, eliminating the need for MLP-based motion modeling. In canonical space, we replace traditional MLP decoders with a structured SH-based rendering head that synthesizes view-dependent color via attention over learned frequency bands improving both interpretability and rendering efficiency. To further enhance fidelity and temporal consistency, we introduce a transformer-guided latent diffusion module that refines the tri-plane and deformation features in a compressed latent space. This generative module denoises scene representations under ambiguous or out-of-distribution (OOD) motion, improving generalization. Our model is trained in two stages: the diffusion module is first pre-trained independently, and then fine-tuned jointly with the full pipeline using a combination of image reconstruction, diffusion denoising, and temporal consistency losses. We demonstrate state-of-the-art results on synthetic benchmarks, surpassing recent methods such as HexPlane and 4D Gaussian Splatting in visual quality, temporal coherence, and robustness to sparse-view dynamic inputs.

Figures

Figures reproduced from arXiv: 2505.16535 by the authors.

Figure 1
Figure 1. Visualization of our 4D scene reconstruction across time. From left to right, we show reconstructed frames of a dynamic subject at increasing motion timestamps. The rightmost frame corresponds to the canonical configuration into which all motion is explicitly warped for consistent appearance modeling. Our method preserves structural integrity and appearance fidelity over time, even under significant non-rigid deform… view at source ↗
Figure 2
Figure 2. Overview of our dynamic scene reconstruction framework. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Reconstruction quality under varying input sparsity. We compare PSNR values for our method, HexPlane [1], and 4D Gaussian Splatting [24] across increasing numbers of input views (3, 5, 10, 20). Our method retains high fidelity even under extreme view sparsity, demonstrating strong generalization and robustness to limited observations. Pipeline Summary. Our architecture modularizes the key components of 4D reconstruc… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison on the Jumping Jacks, Stand Up, and Lego scenes from the D-NeRF benchmark. Our method produces sharper, more temporally stable reconstructions compared to HexPlane [1] and 4D Gaussian Splatting [24] [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 22 canonical work pages

  1. [1]

    Hexplane: A fast representa- tion for dynamic scenes

    Ang Cao and Justin Johnson. Hexplane: A fast representa- tion for dynamic scenes. CVPR, 2023. 2, 3, 5, 6

    work page 2023

  2. [2]

    Chan, Connor Z

    Eric R. Chan, Connor Z. Lin, Matthew A. Chan, Koki Nagano, Boxiao Pan, Shalini De Mello, Orazio Gallo, Leonidas Guibas, Jonathan Tremblay, Sameh Khamis, Tero Karras, and Gordon Wetzstein. Efficient geometry-aware 3D generative adversarial networks. In arXiv, 2021. 2, 3

    work page 2021

  3. [3]

    Fast dynamic radiance fields with time-aware neural voxels

    Jiemin Fang, Taoran Yi, Xinggang Wang, Lingxi Xie, Xi- aopeng Zhang, Wenyu Liu, Matthias Nießner, and Qi Tian. Fast dynamic radiance fields with time-aware neural voxels. In SIGGRAPH Asia 2022 Conference Papers, 2022. 2

    work page 2022

  4. [4]

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

    Sara Fridovich-Keil, Giacomo Meanti, Frederik Rahbæk Warburg, Benjamin Recht, and Angjoo Kanazawa. K-planes: Explicit radiance fields in space, time, and appearance. In CVPR, 2023. 2, 3

    work page 2023

  5. [5]

    Multiple View Ge- ometry in Computer Vision

    Richard Hartley and Andrew Zisserman. Multiple View Ge- ometry in Computer Vision . Cambridge University Press, 2nd edition, 2003. 2

    work page 2003

  6. [6]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. In NeurIPS, 2020. 2

    2020

  7. [7]

    3d gaussian splatting for real-time radiance field rendering

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics, 42 (4), 2023. 3, 4

    work page 2023

  8. [8]

    Light field rendering

    Marc Levoy and Pat Hanrahan. Light field rendering. In ACM SIGGRAPH, 1996. 2

    work page 1996

Show all 27 references
  1. [9]

    Magic3d: High-resolution text-to-3d content creation, 2023

    Chen-Hsuan Lin, Jun Gao, Luming Tang, Towaki Takikawa, Xiaohui Zeng, Xun Huang, Karsten Kreis, Sanja Fidler, Ming-Yu Liu, and Tsung-Yi Lin. Magic3d: High-resolution text-to-3d content creation, 2023. 2

    2023

  2. [10]

    Ricardo Martin-Brualla, Noha Radwan, Mehdi S. M. Sajjadi, Jonathan T. Barron, Alexey Dosovitskiy, and Daniel Duck- worth. NeRF in the Wild: Neural Radiance Fields for Un- constrained Photo Collections. In CVPR, 2021. 2

    2021

  3. [11]

    Occupancy networks: Learning 3d reconstruction in function space

    Lars Mescheder, Michael Oechsle, Michael Niemeyer, Se- bastian Nowozin, and Andreas Geiger. Occupancy networks: Learning 3d reconstruction in function space. In CVPR,

  4. [12]

    Nerf: Representing scenes as neural radiance fields for view syn- thesis

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis. In ECCV, 2020. 2

    2020

  5. [13]

    Deepsdf: Learning continuous signed distance functions for shape representation

    Jeong Joon Park et al. Deepsdf: Learning continuous signed distance functions for shape representation. In CVPR, 2019. 2

    2019

  6. [14]

    Barron, Sofien Bouaziz, Dan B Goldman, Steven M

    Keunhong Park, Utkarsh Sinha, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Steven M. Seitz, and Ricardo Martin-Brualla. Nerfies: Deformable neural radiance fields. ICCV, 2021. 2, 3

    2021

  7. [15]

    Barron, Sofien Bouaziz, Dan B Goldman, Ricardo Martin- Brualla, and Steven M

    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. ACM Trans. Graph., 40(6), 2021. 2, 3

    2021

  8. [16]

    Barron, and Ben Milden- hall

    Ben Poole, Ajay Jain, Jonathan T. Barron, and Ben Milden- hall. Dreamfusion: Text-to-3d using 2d diffusion, 2022. 2, 3

    2022

  9. [17]

    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 CVPR, 2022. 2, 5, 6

    2022

  10. [18]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In CVPR, 2022. 2, 3

    2022

  11. [19]

    Photo tourism: Exploring photo collections in 3d

    Steven M Seitz, Brian Curless, James Diebel, Daniel Scharstein, and Richard Szeliski. Photo tourism: Exploring photo collections in 3d. In SIGGRAPH, 2006. 2

    2006

  12. [20]

    Scene representation networks: Continuous 3d- structure-aware neural scene representations

    Vincent Sitzmann, Michael Zollh ¨ofer, and Gordon Wet- zstein. Scene representation networks: Continuous 3d- structure-aware neural scene representations. In Advances in Neural Information Processing Systems, 2019. 2

    2019

  13. [21]

    Denois- ing diffusion implicit models, 2022

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models, 2022. 2, 5

    2022

  14. [22]

    Score-guided diffusion for 3d human recovery

    Anastasis Stathopoulos, Ligong Han, and Dimitris Metaxas. Score-guided diffusion for 3d human recovery. In CVPR,

  15. [23]

    Yeh, and Greg Shakhnarovich

    Haochen Wang, Xiaodan Du, Jiahao Li, Raymond A. Yeh, and Greg Shakhnarovich. Score jacobian chaining: Lift- ing pretrained 2d diffusion models for 3d generation. arXiv preprint arXiv:2212.00774, 2022. 2

    2022 arXiv

  16. [24]

    4d gaussian splatting for real-time dynamic scene render- ing

    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 render- ing. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR), pages 2...

    2024

  17. [25]

    Plenoxels: Radiance fields without neural networks, 2021

    Alex Yu, Sara Fridovich-Keil, Matthew Tancik, Qinhong Chen, Benjamin Recht, and Angjoo Kanazawa. Plenoxels: Radiance fields without neural networks, 2021. 2, 3, 4

    2021

  18. [26]

    Plenoctrees for real-time rendering of neural radiance fields, 2021

    Alex Yu, Ruilong Li, Matthew Tancik, Hao Li, Ren Ng, and Angjoo Kanazawa. Plenoctrees for real-time rendering of neural radiance fields, 2021. 4

    2021

  19. [27]

    Point cloud pre-training with diffusion models, 2023

    Xiao Zheng, Xiaoshui Huang, Guofeng Mei, Yuenan Hou, Zhaoyang Lyu, Bo Dai, Wanli Ouyang, and Yongshun Gong. Point cloud pre-training with diffusion models, 2023. 2

    2023

Pith tools

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