Pith. sign in

REVIEW 5 major objections 4 minor 48 references

3D$^2$-Actor: Learning Pose-Conditioned 3D-Aware Denoiser for Realistic Gaussian Avatar Modeling

T0 review · 5 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read 3D2-Actor claims that a pose-conditioned 3D-aware denoiser, which interleaves 2D diffusion steps with 3D Gaussian rectification, can synthesize realistic and animatable human avatars from sparse multi-view video and generalize to novel…

desk verdict A genuinely new interleaving of 2D diffusion and 3DGS for avatars, but test-time hyperparameter search and an unaddressed train-test gap in the 3D rectifier undermine the reported numbers. read the letter →

arxiv 2412.11599 v1 pith:MMPBW2RP submitted 2024-12-16 cs.CV

Zichen Tang , Hongyu Yang , Hanchen Zhang , Jiaxin Chen , Di Huang This is my paper · ORCID
classification cs.CV
keywords 3DhumanavatarGaussiansplattingdiffusionmodelpose-conditionedgenerationnovelposesynthesistemporalconsistencymulti-viewreconstructionanimatable
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

3D2-Actor claims that animatable human avatars can be synthesized by treating reconstruction as a pose-conditioned denoising problem: start from pure noise, alternate a 2D diffusion denoiser that generates detailed multi-view images with a 3D Gaussian rectifier that forces the images to be consistent views of a single 3D body, and condition both stages on SMPL body pose. The paper argues this generative route handles the ill-posed mapping from pose to appearance better than traditional deformation-based numerical solvers, capturing stochastic high-frequency details such as clothing wrinkles and generalizing to poses not seen in training. If correct, the method would give a practical way to create realistic, animatable avatars from sparse multi-view RGB video, with perceptual quality that outperforms existing deformation-based methods even when pixel-level accuracy is slightly lower.

What carries the argument

The key mechanism is an alternating denoising loop composed of a 2D U-Net denoiser and a 3D Gaussian rectifier. The 2D denoiser uses SFT layers modulated by body segmentation maps to predict clean images from noisy ones, adding high-frequency detail; the 3D rectifier takes those clean images, extracts pixel-aligned features, and reconstructs a set of 3D Gaussians whose positions are expressed as local coordinate quaternions $\xi = (\lambda_1,\lambda_2,\lambda_3,m)$ against the posed SMPL mesh, where the $\lambda$'s are barycentric coordinates on the nearest triangle and $m$ is the distance along the triangle normal. This representation lets the pipeline propagate Gaussian positions from one frame to the next simply by re-evaluating the local coordinates on the new mesh, which is the foundation of both the two-stage projection and the temporal consistency sampling strategy.

What would settle it

A concrete test would be to feed the 3D rectifier with real ground-truth frames and with the 2D denoiser's generated clean frames for the same poses, then compare the final rendered image errors; if rendering error is systematically higher for denoiser-generated inputs, the train-test distribution shift is confirmed as the limiting factor.

Watch

Extended reading notes

Core claim

The central claim is that interleaving 2D denoising with 3D rectification yields a 3D-aware denoiser that can generate realistic and animatable human avatars. The 3D rectifier reconstructs 3D Gaussians from clean multi-view images produced by the 2D denoiser using a two-stage projection onto pixel-aligned features, and Gaussian positions are stored in a mesh-based local coordinate representation tied to the posed SMPL mesh. The rectifier's rendered images are fed back into the denoising loop through the DDIM sampling trick, allowing the two modules to refine each other. On four ZJU-MoCap sequences, the approach reports lower LPIPS and FID than ARAH and PoseVocab for single-frame novel pose synthesis, indicating better perceptual fidelity and generative quality, though PSNR is not always the highest. The paper also introduces an inter-frame Gaussian consistency sampling strategy that carries local coordinates from the previous frame to the current one, reducing flicker in generated videos without training a temporal module.

Load-bearing premise

The load-bearing premise is that the 3D rectifier, which is trained on ground-truth clean frames from the dataset, will also work correctly when it is fed clean frames generated by the 2D denoiser at inference time, even though the paper never analyzes or adapts to this distribution shift; if the denoiser's outputs lie far from the training distribution, the reconstructed Gaussians and all downstream renderings degrade.

Editorial extensions

If this is right

  • If the central claim is correct, the proposed 3D-aware denoiser offers a generative alternative to per-frame latent code optimization, producing sharper clothing details on novel poses than ARAH and PoseVocab.
  • The mesh-based local coordinate representation allows Gaussian positions to be re-evaluated on a new SMPL mesh, so the same pipeline can be used for both novel-pose synthesis and video animation without retraining.
  • Inserting 3D rectifying steps only in the early denoising stages, as the method does, concentrates 3D consistency enforcement on large-scale structure while leaving fine texture generation to the later 2D steps.
  • The Gaussian consistency sampling strategy reduces inter-frame flicker in generated videos compared to generating each frame independently, as evidenced by lower adjacent-frame LPIPS and optical flow warp error.

Reading between the lines

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

  • The same interleaved 2D-denoise / 3D-rectify loop could transfer to other articulated objects that have a mesh prior, such as animal bodies or mechanical hands, since the local-coordinate propagation assumes only a deforming mesh.
  • A direct extension would be to fine-tune the 3D rectifier on denoiser outputs, or to add a cycle-consistency loss, which would directly address the train-test distribution shift and potentially close the gap between training and inference behavior.
  • The temporal sampling strategy effectively turns video generation into an incremental denoising problem, a mechanism that could be adapted to other generative video tasks where a 3D or mesh prior is available.
  • Because the scaffold is a fixed SMPL mesh, the method inherits SMPL's limitations; modeling loose or non-rigid clothing would likely require a learned scaffold or a second mesh pass.
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 3D^2-Actor, a pose-conditioned 3D-aware denoiser that interleaves a 2D diffusion-style denoiser with a Gaussian-based 3D rectifier for reconstructing and animating human avatars from sparse multi-view RGB video. The 2D denoiser is conditioned on SMPL segmentation maps to refine local details, while the 3D rectifier reconstructs 3D Gaussians from clean multi-view images using a two-stage projection strategy with a mesh-based local coordinate representation. The authors also introduce an inter-frame Gaussian consistency sampling strategy for temporally coherent video synthesis. Experiments on ZJU-MoCap report improved LPIPS and FID over ARAH and PoseVocab in single-frame novel pose synthesis, with ablations for the 2D/3D components, the hyperparameters (tsplit, k), and the local coordinate representation. Code is released on GitHub.

Significance. If the results hold, the paper makes a useful contribution by demonstrating a viable way to combine 2D diffusion refinement with explicit 3D Gaussian reconstruction for animatable avatars, addressing the multi-view consistency problem that pure 2D diffusion models face. The proposed local-coordinate representation and the inter-frame sampling strategy are interesting and are supported by ablations and supplementary video metrics. The authors also provide code, which aids reproducibility. However, the empirical validation is weakened by test-set hyperparameter selection, a train-test distribution gap in the 3D rectifier, the absence of recent Gaussian-avatar baselines, and a lack of repeated-run variance reporting. These issues do not invalidate the core pipeline but they do mean the reported quantitative gains should be interpreted with caution until the evaluation is made fairer and more robust.

major comments (5)
  1. [Single-Frame Novel Pose Synthesis, Table 3] The authors state that for novel pose synthesis they 'search for the best tsplit, kpair for each ID' and the supplementary tables show results across a grid of (tsplit, k). If this search is performed on the test frames, the reported numbers in Table 1 are partly fitted to the test set, which makes the comparison against ARAH and PoseVocab unfair. This concern is reinforced by the supplementary statement that results for tsplit > 300 are withheld 'as the PSNR value is not ideal enough.' The hyperparameters should be selected on a validation split derived from training frames, or the method should be evaluated under fixed settings across all sequences, with the test-grid results reported only as sensitivity analysis. This is load-bearing because the headline LPIPS/FID gains depend on this per-sequence test-set selection.
  2. [Training objective, Eq. (8)] There is a train-test input distribution gap for the 3D rectifier G3D. Eq. (8) trains G3D on ground-truth dataset frames I_f^(1:N), but at inference every call to G3D inside the 3D-aware denoising loop receives clean images synthesized by the 2D denoiser D2D, not dataset frames. The paper provides no fine-tuning of G3D on D2D outputs, no domain-adaptation step, and no quantitative analysis of how far D2D-generated clean images are from the training distribution. The ablation in Table 2 is telling: Ours-3D, which feeds a D2D output into a single 3D rectifier, achieves FID 115.8/79.7 on sequences 313/315, versus 19.5/20.2 for the full model. This shows that G3D alone is not robust to generated inputs and suggests the rectifier's behavior on D2D outputs is fragile. The authors should analyze this shift, adapt G3D accordingly, or at minimum measure the distributional gap and its effect on the final renderings.
  3. [Training objective, Eq. (10)] The training objective for the 2D denoiser is under-specified. Eq. (10) is written as L2D = ||D2D(I_f, S_f, t) - I_f||^2 with I_f described as a clean video frame, yet the method section says the denoiser takes noisy images as input. The equation omits the noise-perturbation term, so it is unclear whether the input to D2D is noised during training, how the timestep t is sampled, and whether the objective is to predict the clean image from a noised observation. Similarly, Eq. (8) does not show how the timestep t that appears in Eq. (6) is sampled for the 3D rectifier. These details are essential for reproducibility and should be stated precisely.
  4. [Baselines, Table 1] The quantitative comparison includes only ARAH and PoseVocab. Both are reasonable earlier baselines, but the related work section itself cites recent Gaussian-based avatar methods (GauHuman, 3DGS-Avatar, Animatable Gaussians, Human101, GEA) that report ZJU-MoCap results. Omitting all of these undercuts the claim that 3D^2-Actor achieves state-of-the-art performance. The authors should add comparisons against at least the most relevant recent Gaussian avatar methods with the same train/test split, or explicitly justify why those methods are excluded. This is a key part of the paper's central claim of high-fidelity avatar modeling.
  5. [Evaluation protocol, Tables 1-3] The paper reports no variance across stochastic runs. Since the method is generative and the reported metrics (PSNR, LPIPS, FID) are computed on images produced through a stochastic denoising process, single-run numbers can be misleading, especially when the differences between methods are modest (e.g., LPIPS 0.080 vs 0.090 in Table 1, or FID 19.5 vs 20.2). The authors should report means and standard deviations over multiple seeds or multiple initial noise draws, or at least verify that the reported differences are consistent across runs. Without this, the statistical reliability of the claimed improvements cannot be assessed.
minor comments (4)
  1. [Method, first paragraph] There is a typo: 'The details of the the 2D denoiser' should read 'The details of the 2D denoiser.'
  2. [Single-Frame Novel Pose Synthesis] The phrase 'tsplit, kpair' is unclear; it should be written as 'tsplit and k' or 'the best (tsplit, k) pair.'
  3. [Preliminary, Eq. (4)] The notation 'covariance matrix Σ ∈ R^7' is imprecise: a 3D Gaussian covariance is a 3×3 matrix, and the paper should state that the 7 parameters correspond to a standard 3DGS parameterization (e.g., a quaternion and scale).
  4. [Ablation Studies, Table 2] The 'Ours-3D' ablation is described as retaining the initial 3D rectifier but omitting later 2D or 3D steps; the paper should clarify whether the input to that rectifier is the output of the first 2D denoising step or a ground-truth image, since this directly relates to the train-test shift discussed in the major comments.

Circularity Check

1 steps flagged · score 2.0 of 10

No derived circularity; the main pipeline is an engineered composition with independent content, but the reported test metrics are mildly biased by per-sequence search of tsplit and k on the test IDs.

  1. fitted input called prediction [Single-Frame Novel Pose Synthesis, Quantitative Results (Table 1 and surrounding text)]
    "When performing novel pose synthesis on different IDs, we search for the best tsplit, kpair for each ID regarding their various clothes wrinkles and action dynamics."

    The paper reports FID and LPIPS on the test frames of each ZJU-MoCap sequence after choosing tsplit and k per sequence. Because the same test frames are used both to select these hyperparameters and to compute the reported metrics, the Table 1 numbers are partly the result of test-time selection rather than an evaluation of a fixed, pre-specified pipeline. The claim that the method achieves the best or second-best results across the four sequences is therefore somewhat forced by the search. This is a mild form of fitted input called prediction, though it does not make the architecture's internal derivation circular.

full rationale

The paper does not contain a formal derivation chain that would allow a classic circularity failure: 3D2-Actor is an engineered pipeline that interleaves a learned 2D denoiser and a learned 3D Gaussian rectifier. The 3D rectifier is trained with Eq. 8 to reconstruct Gaussians from ground-truth frames and later receives clean images produced by the 2D denoiser at inference; this is a real train/test distribution shift, not a self-referential reduction, and the Ours-3D ablation (FID 115.8/79.7 on sequences 313/315 versus 19.5/20.2 for the full model) shows the rectifier alone is not realistic. Similarly, Eq. 10 trains the 2D denoiser on clean frames to reproduce clean frames, while inference applies it to noised inputs; this is an under-specified training objective and a robustness concern, but it is not a case where the prediction is equivalent to its input by construction. There are no load-bearing self-citations, no imported uniqueness theorems, and no ansatz smuggled in through prior work by the same authors. The only issue relevant to circularity is the per-sequence search over tsplit and k on the test IDs before reporting Table 1, which introduces mild selection bias in the reported numbers. That does not undermine the independent content of the method itself, so the circularity score is low.

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

The main fitted quantities are tsplit and k, selected per sequence on the test set; the method introduces no new physical entities. The axioms are standard tools and dataset assumptions.

free parameters (4)
  • tsplit = 300 (searched over {100,200,300})
    Split timestep between early 3D rectifying steps and late 2D refinement; authors search for the best value per sequence and report the best metrics.
  • k = 2 to 5 per sequence
    Number of inserted 3D rectifying steps; authors state the optimal k differs for various sequences, i.e. tuned per ID.
  • number of Gaussians n = 373056
    Number of Gaussians sampled on SMPL mesh; chosen without sensitivity analysis.
  • loss weights lambda_rgb, lambda_mask = not specified
    Loss weights in Eq. 9 are not reported, so they are effectively free hand-chosen constants.
assumptions (4)
  • domain assumption SMPL mesh provides sufficiently accurate body geometry and segmentation for the local coordinate projection.
    The entire two-stage projection and local coordinate representation assumes the posed SMPL model aligns with the actor; poor registration would corrupt Gaussian positions.
  • domain assumption 3D Gaussian Splatting is a valid differentiable renderer for this task.
    Equation (4) is taken from Kerbl et al. 2023; the method relies on its gradients and rendering quality.
  • standard math The standard diffusion forward/inverse process (Eqs. 1-3) and DDIM sampling (Eq. 5) are valid.
    The method builds on established diffusion theory.
  • domain assumption ZJU-MoCap ground-truth images and the 80/20 train/test split are reliable.
    All metrics are computed against this dataset; the paper does not validate on other data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of 3D$^2$-Actor: Learning Pose-Conditioned 3D-Aware Denoiser for Realistic Gaussian Avatar Modeling." pith.science (2026). https://pith.science/paper/MMPBW2RP

@misc{pith2026241211599,
  author       = {Pith},
  title        = {Pith review of: 3D$^2$-Actor: Learning Pose-Conditioned 3D-Aware Denoiser for Realistic Gaussian Avatar Modeling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MMPBW2RP}},
  note         = {Machine review of arXiv:2412.11599}
}
abstract

Advancements in neural implicit representations and differentiable rendering have markedly improved the ability to learn animatable 3D avatars from sparse multi-view RGB videos. However, current methods that map observation space to canonical space often face challenges in capturing pose-dependent details and generalizing to novel poses. While diffusion models have demonstrated remarkable zero-shot capabilities in 2D image generation, their potential for creating animatable 3D avatars from 2D inputs remains underexplored. In this work, we introduce 3D$^2$-Actor, a novel approach featuring a pose-conditioned 3D-aware human modeling pipeline that integrates iterative 2D denoising and 3D rectifying steps. The 2D denoiser, guided by pose cues, generates detailed multi-view images that provide the rich feature set necessary for high-fidelity 3D reconstruction and pose rendering. Complementing this, our Gaussian-based 3D rectifier renders images with enhanced 3D consistency through a two-stage projection strategy and a novel local coordinate representation. Additionally, we propose an innovative sampling strategy to ensure smooth temporal continuity across frames in video synthesis. Our method effectively addresses the limitations of traditional numerical solutions in handling ill-posed mappings, producing realistic and animatable 3D human avatars. Experimental results demonstrate that 3D$^2$-Actor excels in high-fidelity avatar modeling and robustly generalizes to novel poses. Code is available at: https://github.com/silence-tang/GaussianActor.

Figures

Figures reproduced from arXiv: 2412.11599 by the authors.

Figure 1
Figure 1. Illustration of the 3D-aware denoising process. (a) The 3D-aware denoising pipeline consists of interlaced 2D and [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. An illustration of the inter-frame Gaussian consistency sampling strategy for improving temporal continuity. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 4
Figure 4. Consecutive frame generation results. Top row [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Novel pose synthesis results with different designs [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 1
Figure 1. Figure 1: Illustration of our pixel-aligned feature extraction network. (a) Only the contracting path of the U-Net is displayed. [PITH_FULL_IMAGE:figures/full_fig_p011_1.png]
Figure 2
Figure 2. Figure 2: Additional novel view synthesis results generated with different design of our 3D-aware denoising process. “Ours [PITH_FULL_IMAGE:figures/full_fig_p011_2.png]
Figure 3
Figure 3. Figure 3: Additional novel view synthesis results generated [PITH_FULL_IMAGE:figures/full_fig_p012_3.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 20 canonical work pages

  1. [1]

    J.; and Guerrero, P

    Anciukevi c ius, T.; Xu, Z.; Fisher, M.; Henderson, P.; Bilen, H.; Mitra, N. J.; and Guerrero, P. 2023. Renderdiffusion: Image diffusion for 3d reconstruction, inpainting and generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12608--12618

    work page 2023

  2. [2]

    J.; Loper, M.; and Romero, J

    Bogo, F.; Black, M. J.; Loper, M.; and Romero, J. 2015. Detailed full-body reconstructions of moving people from monocular RGB-D sequences. In Proceedings of the IEEE international conference on computer vision, 2300--2308

    work page 2015

  3. [3]

    Chen, H.; Gu, J.; Chen, A.; Tian, W.; Tu, Z.; Liu, L.; and Su, H. 2023. Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction. In ICCV

    2023

  4. [4]

    Collet, A.; Chuang, M.; Sweeney, P.; Gillett, D.; Evseev, D.; Calabrese, D.; Hoppe, H.; Kirk, A.; and Sullivan, S. 2015. High-quality streamable free-viewpoint video. ACM Transactions on Graphics (ToG), 34(4): 1--13

    work page 2015

  5. [5]

    R.; Kowdle, A.; Escolano, S

    Dou, M.; Khamis, S.; Degtyarev, Y.; Davidson, P.; Fanello, S. R.; Kowdle, A.; Escolano, S. O.; Rhemann, C.; Kim, D.; Taylor, J.; et al. 2016. Fusion4d: Real-time performance capture of challenging scenes. ACM Transactions on Graphics (ToG), 35(4): 1--13

    work page 2016

  6. [6]

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

    2021

  7. [7]

    Gupta, A.; Xiong, W.; Nie, Y.; Jones, I.; and O g uz, B. 2023. 3dgen: Triplane latent diffusion for textured mesh generation. arXiv preprint arXiv:2303.05371

    arXiv 2023

  8. [8]

    Heusel, M.; Ramsauer, H.; Unterthiner, T.; Nessler, B.; and Hochreiter, S. 2017. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30

    2017

Show all 48 references
  1. [9]

    Ho, J.; Jain, A.; and Abbeel, P. 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33: 6840--6851

    2020

  2. [10]

    Hu, S.; and Liu, Z. 2023. Gauhuman: Articulated gaussian splatting from monocular human videos. arXiv preprint arXiv:2312.02973

    2023 arXiv

  3. [11]

    Huang, Y.; Wang, J.; Shi, Y.; Tang, B.; Qi, X.; and Zhang, L. 2023. Dreamtime: An Improved Optimization Strategy for Diffusion-guided 3D Generation. In The Twelfth International Conference on Learning Representations

    2023

  4. [12]

    Jiang, Y.; Liao, Q.; Li, X.; Ma, L.; Zhang, Q.; Zhang, C.; Lu, Z.; and Shan, Y. 2024. UV Gaussians: Joint Learning of Mesh Deformation and Gaussian Textures for Human Avatar Modeling. arXiv preprint arXiv:2403.11589

    2024 arXiv

  5. [13]

    Jung, H.; Brasch, N.; Song, J.; Perez-Pellitero, E.; Zhou, Y.; Li, Z.; Navab, N.; and Busam, B. 2023. Deformable 3d gaussian splatting for animatable human avatars. arXiv preprint arXiv:2312.15059

    2023 arXiv

  6. [14]

    Karnewar, A.; Vedaldi, A.; Novotny, D.; and Mitra, N. J. 2023. Holodiffusion: Training a 3D diffusion model using 2D images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18423--18433

    2023

  7. [15]

    Kerbl, B.; Kopanas, G.; Leimk \"u hler, T.; and Drettakis, G. 2023. 3D Gaussian Splatting for Real-Time Radiance Field Rendering. ACM Transactions on Graphics, 42(4)

    2023

  8. [16]

    Li, M.; Tao, J.; Yang, Z.; and Yang, Y. 2023 a . Human101: Training 100+ fps human gaussians in 100s from 1 view. arXiv preprint arXiv:2312.15258

    2023 arXiv

  9. [17]

    Li, R.; Tanke, J.; Vo, M.; Zollh \"o fer, M.; Gall, J.; Kanazawa, A.; and Lassner, C. 2022. Tava: Template-free animatable volumetric actors. In European Conference on Computer Vision, 419--436. Springer

    2022

  10. [18]

    Li, Z.; Zheng, Z.; Liu, Y.; Zhou, B.; and Liu, Y. 2023 b . PoseVocab: Learning Joint-structured Pose Embeddings for Human Avatar Modeling. In ACM SIGGRAPH Conference Proceedings

    2023

  11. [19]

    Li, Z.; Zheng, Z.; Wang, L.; and Liu, Y. 2024. Animatable gaussians: Learning pose-dependent gaussian maps for high-fidelity human avatar modeling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 19711--19722

    2024

  12. [20]

    Liu, R.; Wu, R.; Van Hoorick, B.; Tokmakov, P.; Zakharov, S.; and Vondrick, C. 2023. Zero-1-to-3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9298--9309

    2023

  13. [21]

    Liu, X.; Wu, C.; Liu, X.; Liu, J.; Wu, J.; Zhao, C.; Feng, H.; Ding, E.; and Wang, J. 2024 a . GEA: Reconstructing Expressive 3D Gaussian Avatar from Monocular Video. arXiv preprint arXiv:2402.16607

    2024 arXiv

  14. [22]

    Liu, Y.; Lin, C.; Zeng, Z.; Long, X.; Liu, L.; Komura, T.; and Wang, W. 2024 b . SyncDreamer: Generating Multiview-consistent Images from a Single-view Image. In The Twelfth International Conference on Learning Representations

    2024

  15. [23]

    Lombardi, S.; Simon, T.; Saragih, J.; Schwartz, G.; Lehrmann, A.; and Sheikh, Y. 2019. Neural volumes: Learning dynamic renderable volumes from images. arXiv preprint arXiv:1906.07751

    2019 arXiv

  16. [24]

    Loper, M.; Mahmood, N.; Romero, J.; Pons-Moll, G.; and Black, M. J. 2015. SMPL: A Skinned Multi-Person Linear Model. ACM Transactions on Graphics, 34(6)

    2015

  17. [25]

    P.; Tancik, M.; Barron, J

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

    2021

  18. [26]

    R.; Kontschieder, P.; and Nie ner, M

    M \"u ller, N.; Siddiqui, Y.; Porzi, L.; Bulo, S. R.; Kontschieder, P.; and Nie ner, M. 2023. Diffrf: Rendering-guided 3d radiance field diffusion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4328--4338

    2023

  19. [27]

    Nichol, A.; Jun, H.; Dhariwal, P.; Mishkin, P.; and Chen, M. 2022. Point-e: A system for generating 3d point clouds from complex prompts. arXiv preprint arXiv:2212.08751

    2022 arXiv

  20. [28]

    T.; Bouaziz, S.; Goldman, D

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

    2021 arXiv

  21. [29]

    Peng, S.; Dong, J.; Wang, Q.; Zhang, S.; Shuai, Q.; Zhou, X.; and Bao, H. 2021 a . Animatable neural radiance fields for modeling dynamic human bodies. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 14314--14323

    2021

  22. [30]

    Peng, S.; Zhang, Y.; Xu, Y.; Wang, Q.; Shuai, Q.; Bao, H.; and Zhou, X. 2021 b . Neural body: Implicit neural representations with structured latent codes for novel view synthesis of dynamic humans. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog...

    2021

  23. [31]

    T.; Bermano, A

    Po, R.; Yifan, W.; Golyanik, V.; Aberman, K.; Barron, J. T.; Bermano, A. H.; Chan, E. R.; Dekel, T.; Holynski, A.; Kanazawa, A.; et al. 2023. State of the Art on Diffusion Models for Visual Computing. arXiv preprint arXiv:2310.07204

    2023 arXiv

  24. [32]

    Pumarola, A.; Corona, E.; Pons-Moll, G.; and Moreno-Noguer, F. 2021. D-nerf: Neural radiance fields for dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10318--10327

    2021

  25. [33]

    Qian, S.; Kirschstein, T.; Schoneveld, L.; Davoli, D.; Giebenhain, S.; and Nie ner, M. 2023 a . Gaussianavatars: Photorealistic head avatars with rigged 3d gaussians. arXiv preprint arXiv:2312.02069

    2023 arXiv

  26. [34]

    Qian, Z.; Wang, S.; Mihajlovic, M.; Geiger, A.; and Tang, S. 2023 b . 3dgs-avatar: Animatable avatars via deformable 3d gaussian splatting. arXiv preprint arXiv:2312.09228

    2023 arXiv

  27. [35]

    Rombach, R.; Blattmann, A.; Lorenz, D.; Esser, P.; and Ommer, B. 2022. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 10684--10695

    2022

  28. [36]

    Ronneberger, O.; Fischer, P.; and Brox, T. 2015. U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention--MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part II...

    2015

  29. [37]

    Shapiro, A.; Feng, A.; Wang, R.; Li, H.; Bolas, M.; Medioni, G.; and Suma, E. 2014. Rapid avatar capture and simulation using commodity depth sensors. Computer Animation and Virtual Worlds, 25(3-4): 201--211

    2014

  30. [38]

    Sitzmann, V.; Martel, J.; Bergman, A.; Lindell, D.; and Wetzstein, G. 2020. Implicit neural representations with periodic activation functions. Advances in neural information processing systems, 33: 7462--7473

    2020

  31. [39]

    Song, J.; Meng, C.; and Ermon, S. 2020. Denoising Diffusion Implicit Models. In International Conference on Learning Representations

    2020

  32. [40]

    Su, S.-Y.; Bagautdinov, T.; and Rhodin, H. 2023. NPC: Neural Point Characters from Video. In Proceedings of the IEEE/CVF International conference on computer vision, 14795--14805

    2023

  33. [41]

    Su, S.-Y.; Yu, F.; Zollh \"o fer, M.; and Rhodin, H. 2021. A-nerf: Articulated neural radiance fields for learning human shape, appearance, and pose. Advances in Neural Information Processing Systems, 34: 12278--12291

    2021

  34. [42]

    Szymanowicz, S.; Rupprecht, C.; and Vedaldi, A. 2023. Viewset Diffusion: (0-)Image-Conditioned 3D Generative Models from 2D Data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 8863--8873

    2023

  35. [43]

    Wang, P.; Liu, L.; Liu, Y.; Theobalt, C.; Komura, T.; and Wang, W. 2021. NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction. Advances in Neural Information Processing Systems, 34: 27171--27183

    2021

  36. [44]

    Wang, S.; Schwarz, K.; Geiger, A.; and Tang, S. 2022. Arah: Animatable volume rendering of articulated human sdfs. In European conference on computer vision, 1--19. Springer

    2022

  37. [45]

    Wang, X.; Yu, K.; Dong, C.; and Loy, C. C. 2018. Recovering realistic texture in image super-resolution by deep spatial feature transform. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    2018

  38. [46]

    A.; Shechtman, E.; and Wang, O

    Zhang, R.; Isola, P.; Efros, A. A.; Shechtman, E.; and Wang, O. 2018. The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    2018

  39. [47]

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

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  40. [48]

    write newline

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

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.