Pith. sign in

REVIEW 4 major objections 7 minor 79 references

Reconstructing 3D Flow from 2D Data with Diffusion Transformer

T0 review · 4 major / 7 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 A diffusion transformer conditioned on two orthogonal 2D velocity slices can reconstruct the full 3D turbulent flow field, reaching near-simulation accuracy on in-distribution DNS data.

desk verdict A plausible engineering contribution whose headline accuracy number is inflated by the interpolation split; the honest extrapolation result is only marginal, so the paper needs major revision before its claims can be trusted. read the letter →

arxiv 2502.02593 v1 pith:6ZTJCHBR submitted 2024-12-20 cs.CE cs.AIphysics.flu-dyn

Fan Lei This is my paper · ORCID
classification cs.CEcs.AIphysics.flu-dyn
keywords 3Dflowreconstructiondiffusiontransformerplanepositionembeddingparticleimagevelocimetryturbulentchannelwindowattentionvoxelspace
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

This paper tries to establish that a diffusion transformer can recover a full three-dimensional velocity field from just two orthogonal two-dimensional slices of it. If true, experimentalists could get 3D flow data from cheaper 2D particle-image velocimetry instead of 3D setups with multiple cameras, and CFD users could reduce reliance on expensive full-volume simulations. The authors train the model on direct numerical simulation data, condition it on the two slices through several pathways, and add plane position embeddings so the same model can handle slices at any location. On the in-distribution interpolation benchmark their Base model reaches a normalized RMSE of 0.0053 and an SSIM of 0.9997, and the model still produces meaningful flow when one reference plane is withheld.

What carries the argument

The load-bearing identity is the simplified denoising objective $$L_{\text{simple}}(\$\theta$)=\mathbb{E}_{t,S_0,\epsilon}\|\epsilon-\epsilon_\$\theta$(S_t,c(t,P,E_P))\|$_2^{2}$,$$ which trains the transformer to predict the noise added to the 3D volume while conditioned on the 2D planes $P$ and their positions $E_P$. Two mechanisms make this practical: plane position embeddings, which are Fourier features of the normalized plane coefficients $A',B',C',D'$ in the plane equation $A'x+B'y+C'z+D'=0$, and two linear-complexity attention schemes. Window attention groups tokens into $w\times w\times w$ cubes, reducing complexity from $O(L^2)$ to $O(Lw^3)$, and plane attention lets tokens attend only within the $yOz$, $xOz$, and $xOy$ planes. Together they give the model slice-location awareness while keeping 3D token processing affordable.

What would settle it

A concrete experiment: take the INS(INT) test set, add Gaussian noise with standard deviation of 1-5% of the mean velocity to the two input slices, and measure nRMSE; if the error rises sharply from the clean-slice value of 0.0053, the noiseless-slice assumption is load-bearing.

Watch

Extended reading notes

Core claim

The central claim, stated in the authors' terms, is that a voxel-space diffusion transformer can act as a learned prior for turbulent flows: it fills in the unmeasured volume between and beyond two input slices with near-ground-truth accuracy when the flow regime is covered by training data. The conditioning has three streams: the 2D slices are padded and channel-concatenated with the noisy 3D volume, a pretrained image encoder's global and final-layer features are injected through adaptive layer norm and cross-attention, and the normalized plane equation coefficients are Fourier-embedded so the model knows each slice's position. The plane position embedding is what enables reconstruction from arbitrary slice combinations rather than a fixed measurement geometry.

Load-bearing premise

The load-bearing premise is that the two input planes are exact, noiseless voxel slices of the target 3D field; if real PIV data are noisy, downsampled, or misaligned, the model has not been shown to handle it.

Editorial extensions

If this is right

  • On the in-distribution interpolation benchmark, the Base model reaches nRMSE 0.0053, PSNR 51.0158, and SSIM 0.9997, meaning the reconstructed 3D volume is numerically and visually nearly indistinguishable from the DNS ground truth.
  • A single trained model can reconstruct from arbitrary plane combinations because plane position embeddings encode where each slice lies; adding an extra plane does not always improve quality and in some cases slightly hurts it.
  • Window and plane attention speed up training by 27.9-39.3% across model sizes while barely changing accuracy, and for the Large model the efficient attention variant actually outperforms global attention.
  • On extrapolation benchmarks (INS(EXT), CNS), the method remains competitive and its SSIM lead indicates better-preserved small-scale structure, though reconstruction quality degrades with distance from the input planes.
  • When one reference plane is withheld, the diffusion model still produces a meaningful flow field, whereas the strongest non-diffusion baseline produces meaningless noise on that plane.

Reading between the lines

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

  • Inference: because the conditioning slices are exact voxels, the 0.0053 nRMSE result is likely an upper bound on real-world performance; adding PIV-like noise or slight slice misalignment would degrade it, and the size of that degradation is the real test of practical value.
  • Inference: the conditioning uses a pretrained image encoder whose representations were learned on photographs, which is a domain mismatch for multi-channel velocity slices; a physics-specific encoder or one based on local flow statistics might be simpler and equally effective.
  • Inference: the plane position embedding parameterizes any plane intersecting the unit cube, so the same framework could extend to non-axis-aligned cutting planes, non-uniform sensor placements, or time-resolved reconstruction.
  • Inference: the quality-versus-distance results imply an optimal sensor-placement strategy: multiple evenly spaced reference planes give better coverage than many planes clustered near one location, which could directly inform PIV experiment design.
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

4 major / 7 minor

Summary. The paper proposes a diffusion-transformer method for reconstructing a full 3D velocity field from a small set of 2D planar slices. The model is conditioned in three ways: by padding and channel-concatenating the input slices with the volume being denoised, by injecting CLIP image features and plane position embeddings through adaptive layer norm, and by cross-attending to the CLIP encoder output. Global attention is replaced with window attention and plane attention to reduce the cost of processing 3D voxels. Experiments are carried out on DNS data for a turbulent channel flow at Re_tau=180 (INS) and compressible Navier-Stokes data (CNS), plus a geometry dataset, and the model is compared with F-FNO, U-Net-mod, Dil-ResNet, and 2D3DGAN. On the random-split interpolation benchmark INS(INT), the Base model reports nRMSE 0.0053, PSNR 51.0158, and SSIM 0.9997; on the temporally extrapolative split INS(EXT), the same model reports nRMSE 0.1331 versus 0.1421 for F-FNO. The paper concludes that the model can efficiently and accurately reconstruct 3D flow fields from 2D data.

Significance. If the central claim were fully supported, the method would be practically useful for reducing the cost of three-dimensional PIV and CFD studies, and the plane position embedding plus window/plane attention are reasonable architectural contributions. The paper is honest about some limitations, such as the sensitivity to classifier-free guidance, and it provides several ablations, including attention type and patch size. However, the headline accuracy is achieved on a benchmark that the paper itself acknowledges contains near-duplicate training and test snapshots, and on the temporally separated benchmark the gain over F-FNO is marginal. The paper does not ship code, data, or error bars, and diffusion sampling is stochastic, so the reported single point estimates are not enough to establish the comparative claims. The significance is therefore conditional: the architecture is plausible, but the evidence for accurate 3D reconstruction from 2D data is not yet convincing.

major comments (4)
  1. [Section 4.1, Table 2] The INS(INT) benchmark randomly splits DNS snapshots that are only Delta t = 0.07 apart. For a turbulent channel flow at Re_tau=180, this interval is far below the correlation time, so each held-out test snapshot almost certainly has a near-duplicate in the training set. The paper itself states that 'the flow fields at consecutive time steps are similar' (Section 4.1). Because the exact 2D slices are supplied as conditioning, the model can succeed on this split by retrieving a nearly identical training volume rather than by learning a general 2D-to-3D mapping. Consequently the headline numbers nRMSE 0.0053, PSNR 51.0158, SSIM 0.9997 do not measure generalization, and the abstract's unconditional claim that the model 'can efficiently and accurately reconstruct 3D flow fields from 2D data' rests on this contaminated benchmark. The temporally separated INS(EXT) split is the relevant generalization test, and there the Base model achieves only nRMSE 0.1331 versus 0.1421 for F-FNO, which is a marginal improvement. The paper should either report the minimum temporal gap between training and test snapshots in the interpolation split or replace it with a split that guarantees separation, and the conclusions should be scaled to the extrapolation results.
  2. [Section 4.3, Tables 2-6] All reported metrics are single point estimates with no error bars, no multiple seeds, and no statement of the number of diffusion sampling runs averaged. Since the method is a stochastic generative model, the differences between models on the extrapolation benchmarks, which are often only 0.005-0.01 in nRMSE, may be within sampling noise. The paper should report means and standard deviations over multiple training seeds and multiple sampling runs, and should state whether the differences against F-FNO are statistically significant.
  3. [Section 3.3, Section 4.1] The conditioning inputs are exact, noiseless, same-grid DNS slices, while the abstract and introduction motivate the method by PIV experiments, which produce noisy, lower-resolution, and potentially misaligned measurements. The paper does not test robustness to noise, downsampling, or misalignment, so the practical claim that the method can reduce PIV cost is not supported by the experiments. At minimum, a sensitivity study with synthetic noise or degraded slices is needed, or the claims should be restricted to noiseless simulation data.
  4. [Section 4.5, Figure 7] The plane position embedding is one of the three listed contributions, but the ablation in Figure 7 does not isolate its effect: the model is evaluated with three planes at a fixed relative position of -5, and the metrics are plotted against spatial position, but there is no comparison against a model without position embedding, nor against a fixed one-hot or integer plane index. The observed high quality near the reference planes could be driven by the supplied slice content rather than by the learned position encoding. The paper should include a with-versus-without comparison for the plane position embedding across multiple plane configurations.
minor comments (7)
  1. [Abstract] "Here, We propose" has an erroneous capital W; please fix the capitalization and also re-read the abstract for other grammatical slips such as "computationalresources".
  2. [Equation (2)] The norm notation in Equation (2) is malformed: it appears as "∥...∥2^2" without a clear subscript, and the expectation is missing the appropriate parentheses. Please rewrite it as \(\mathbb{E}_{t,S_0,\epsilon}\left[\|\epsilon - \epsilon_\theta(S_t, c(t,P,E_P))\|_2^2\right]\) or an equivalent form.
  3. [Table 2] In the CNS column, the Large* model reports PSNR 25.00220 with an extra decimal place; please make the precision consistent. Also, the text in Section 4.3 says "our Large model may exhibit overfitting," but Table 2 reports results for "Large* (Ours)"; clarify whether the overfitting comment refers to the global-attention Large or the window/plane-attention Large*.
  4. [Table 5] The column heading "Training Relative Promotion (%)" is unclear; replace it with a concrete measure such as training wall-clock time or steps per second, and report absolute values with variance. The phrase "promotion" is not standard in this context.
  5. [Figure 3] The caption "CNSxOy planes where the reference xOy plane is not provided" is confusing; please rephrase to make clear which planes are inputs and which are outputs for each panel.
  6. [Section 4.2] The F-FNO configuration is described as a 24-layer network with 16 Fourier modes and 64 channels, but no citation or appendix section is given for this exact setup. Please provide a reference or a detailed configuration table.
  7. [Table 6] "Pacth Size" is a typo for "Patch Size." Also, when reporting that the model fails to converge for patch size 8, state the training budget and the criterion used to declare non-convergence.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the 3D-from-2D reconstruction is an empirically trained mapping evaluated on held-out data, with no fitted parameter renamed as a prediction and no load-bearing self-citation.

full rationale

The paper contains no derivation step in which a claimed output is defined in terms of its own input, no fitted parameter that is later renamed as a prediction, and no load-bearing self-citation. The method is an empirical supervised/generative mapping: Section 3.1 defines the goal as learning f_theta(P, E_P), and Section 3.3 conditions a diffusion transformer on 2D slices via concatenation, CLIP features, and plane position embeddings. Section 4.3 then evaluates the trained model on held-out DNS test snapshots against independent baselines, so the reported reconstruction quality is an empirical result rather than a consequence of construction. The paper's own note in Section 4.1 that INS(INT) uses randomly split time steps with Delta_t = 0.07, making consecutive flow fields similar, is a statistical-generalization concern about benchmark difficulty, not circularity: the test outputs are still not algebraically forced by the training inputs. The dataset citations [76], [64], and [35] supply external data, and architectural references such as [51], [45], and [70] supply standard building blocks; none of these incorporate the target reconstruction result. No passage asserts a missing proof or a circular step that would alter this verdict. Therefore no specific circular step can be quoted and exhibited, and the appropriate score is 0.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard diffusion-model theory plus two unverified domain assumptions: that slices are exact clean planes of the same grid, and that CLIP features are useful for flow data. The learnable weighting of conditioning streams and hand-chosen architectural scales (patch, window, Fourier multiplier) are free parameters, none of which are tied to an independent physical benchmark.

free parameters (6)
  • patch_size = 4 (main experiments)
    Hand-chosen; ablation (Table 6) shows patch 8 fails to converge and patch 2 trades nRMSE for SSIM.
  • window_size w = 4
    Hand-chosen; determines attention window of w^3 tokens and the claimed 8x complexity reduction.
  • learnable conditioning weights lambda1, lambda2 = both initialized to 1
    Learned from data to weight CLIP feature embedding FP and plane position embedding EP in Eq. (8).
  • Fourier embedding scale = 10,000
    Hand-chosen multiplier applied to normalized plane equation parameters A', B', C', D' in the plane position embedding.
  • inference denoising steps = Not reported
    The number of DDPM sampling steps at test time is not stated, so the reported metrics cannot be reproduced exactly.
  • CLIP image encoder variant = Not specified
    The conditioning depends on a chosen CLIP model, but the paper does not state which checkpoint or resolution is used.
assumptions (4)
  • standard math DDPM reverse process conditioned on c(t,P,EP) learns the flow distribution, with training objective Eq. (2).
    Invoked in Section 3.1; relies on standard denoising diffusion theory where the variational bound reduces to noise prediction.
  • domain assumption The 2D input slices are exact, noiseless voxel planes sampled from the same DNS grid as the target 3D field.
    Used throughout the experiments; the conditioning stream concatenates padded 2D planes with the denoised volume, so exact slice values are provided and the task is an inpainting problem.
  • ad hoc to paper CLIP's image encoder, pre-trained on natural images, provides useful conditioning features for 2D flow velocity fields.
    Section 3.3 extracts CLIP features from flow planes, but no ablation or justification shows that these features help reconstruction.
  • domain assumption DNS data from the cited sources represents the flow regime of interest and is sufficient to learn the 3D-to-2D mapping.
    The method is evaluated only on incompressible channel turbulence at Re_tau=180 and compressible turbulence at Mach 1.0 from [76] and [64]; the transfer to real PIV data is not tested.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reconstructing 3D Flow from 2D Data with Diffusion Transformer." pith.science (2026). https://pith.science/paper/6ZTJCHBR

@misc{pith2026250202593,
  author       = {Pith},
  title        = {Pith review of: Reconstructing 3D Flow from 2D Data with Diffusion Transformer},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6ZTJCHBR}},
  note         = {Machine review of arXiv:2502.02593}
}
read the original abstract

Fluid flow is a widely applied physical problem, crucial in various fields. Due to the highly nonlinear and chaotic nature of fluids, analyzing fluid-related problems is exceptionally challenging. Computational fluid dynamics (CFD) is the best tool for this analysis but involves significant computational resources, especially for 3D simulations, which are slow and resource-intensive. In experimental fluid dynamics, PIV cost increases with dimensionality. Reconstructing 3D flow fields from 2D PIV data could reduce costs and expand application scenarios. Here, We propose a Diffusion Transformer-based method for reconstructing 3D flow fields from 2D flow data. By embedding the positional information of 2D planes into the model, we enable the reconstruction of 3D flow fields from any combination of 2D slices, enhancing flexibility. We replace global attention with window and plane attention to reduce computational costs associated with higher dimensions without compromising performance. Our experiments demonstrate that our model can efficiently and accurately reconstruct 3D flow fields from 2D data, producing realistic results.

Figures

Figures reproduced from arXiv: 2502.02593 by the authors.

Figure 1
Figure 1. Pipeline and architecture. We use the Diffuse Transformer to reconstruct 3D flow from 2D flow. Top: The 2D flow is fed into the model through three paths: 1) it is padded and concatenated with the input; 2) features are extracted using CLIP, combined with the timestep embedding and plane position embedding, and used as conditioning; and 3) the final layer output of the CLIP image encoder is fed into the model via cr… view at source ↗
Figure 2
Figure 2. Visual explanation of different attention. Each grid of the cube represents an input token, and only tokens of the same color can attend to each other. (a) Global attention. (b) Window attention. (c) Plane attention: from left to right, yOz plane attention, xOz plane attention, and xOy plane attention. 3.3 Diffusion Transformer for 3D Flow Field Reconstruc￾tion Inspired by recent works [45, 51, 7], we use a diffusio… view at source ↗
Figure 3
Figure 3. Visualization of results from different models on the INS(INT) (top [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Relative positions to the center of the 3D flow field versus three test [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Relative positions to the center of the 3D flow field versus three test [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Visualization of reconstructing the 3D flow field around geometries. [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: Relative positions to the center of the 3D flow field versus three test [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

79 extracted references · 49 canonical work pages

  1. [1]

    Scattering particle characteristics and their effect on pulsed laser measurements of fluid flow: speckle velocimetry vs particle image velocimetry

    Ronald J Adrian. Scattering particle characteristics and their effect on pulsed laser measurements of fluid flow: speckle velocimetry vs particle image velocimetry. Applied optics, 23(11):1690–1691, 1984

    work page 1984

  2. [2]

    Neural operator: Graph kernel network for partial differential equations

    Anima Anandkumar, Kamyar Azizzadenesheli, Kaushik Bhattacharya, Nikola Kovachki, Zongyi Li, Burigede Liu, and Andrew Stuart. Neural operator: Graph kernel network for partial differential equations. InICLR 2020 Workshop on Integration of Deep Neural Models and Differential Equations, 2020

    work page 2020

  3. [3]

    All are worth words: A vit backbone for diffusion models

    Fan Bao, Shen Nie, Kaiwen Xue, Yue Cao, Chongxuan Li, Hang Su, and Jun Zhu. All are worth words: A vit backbone for diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 22669–22679, 2023

    2023

  4. [4]

    One transformer fits all distributions in multi-modal diffusion at scale

    Fan Bao, Shen Nie, Kaiwen Xue, Chongxuan Li, Shi Pu, Yaole Wang, Gang Yue, Yue Cao, Hang Su, and Jun Zhu. One transformer fits all distributions in multi-modal diffusion at scale. InInternational Conference on Machine Learning, pages 1692–1717. PMLR, 2023

    2023

  5. [5]

    Machine learning for fluid mechanics.Annual review of fluid mechanics, 52(1):477–508, 2020

    Steven L Brunton, Bernd R Noack, and Petros Koumoutsakos. Machine learning for fluid mechanics.Annual review of fluid mechanics, 52(1):477–508, 2020

    2020

  6. [6]

    Mo- tion2vecsets: 4d latent vector set diffusion for non-rigid shape reconstruction and tracking

    Wei Cao, Chang Luo, Biao Zhang, Matthias Nießner, and Jiapeng Tang. Mo- tion2vecsets: 4d latent vector set diffusion for non-rigid shape reconstruction and tracking. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20496–20506, 2024

    work page 2024

  7. [7]

    Pixart-alpha: Fast training of diffusion transformer for photorealistic text-to-image synthesis

    Junsong Chen, YU Jincheng, GE Chongjian, Lewei Yao, Enze Xie, Zhongdao Wang, James Kwok, Ping Luo, Huchuan Lu, and Zhenguo Li. Pixart-alpha: Fast training of diffusion transformer for photorealistic text-to-image synthesis. InThe Twelfth International Conference on Learning Representations

    work page

  8. [8]

    Generative pretraining from pixels

    Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, and Ilya Sutskever. Generative pretraining from pixels. InInternational conference on machine learning, pages 1691–1703. PMLR, 2020

    2020

Show all 79 references
  1. [9]

    Nerdi: Single-view nerf synthesis with language- guided diffusion as general image priors

    Congyue Deng, Chiyu Jiang, Charles R Qi, Xinchen Yan, Yin Zhou, Leonidas Guibas, Dragomir Anguelov, et al. Nerdi: Single-view nerf synthesis with language- guided diffusion as general image priors. InProceedings of the IEEE/CVF confer- ence on computer vision and pattern recog...

    2023

  2. [10]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021

    2021

  3. [11]

    Nice: Non-linear independent components estimation

    Laurent Dinh, David Krueger, and Yoshua Bengio. Nice: Non-linear independent components estimation. arXiv preprint arXiv:1410.8516, 2014

    2014 arXiv

  4. [12]

    Generalization capabilities of conditional gan for turbulent flow under changes of geometry.arXiv preprint arXiv:2302.09945, 2023

    Claudia Drygala, Francesca di Mare, and Hanno Gottschalk. Generalization capabilities of conditional gan for turbulent flow under changes of geometry.arXiv preprint arXiv:2302.09945, 2023. 18

    2023 arXiv

  5. [13]

    Turbulence modeling in the age of data.Annual review of fluid mechanics, 51(1):357–377, 2019

    Karthik Duraisamy, Gianluca Iaccarino, and Heng Xiao. Turbulence modeling in the age of data.Annual review of fluid mechanics, 51(1):357–377, 2019

    2019

  6. [14]

    Reinforcement learning for bluff body active flow control in experi- ments and simulations.Proceedings of the National Academy of Sciences, 117(42): 26091–26098, 2020

    Dixia Fan, Liu Yang, Zhicheng Wang, Michael S Triantafyllou, and George Em Karniadakis. Reinforcement learning for bluff body active flow control in experi- ments and simulations.Proceedings of the National Academy of Sciences, 117(42): 26091–26098, 2020

    2020

  7. [15]

    3d shape induction from 2d views of multiple objects

    Matheus Gadelha, Subhransu Maji, and Rui Wang. 3d shape induction from 2d views of multiple objects. In2017 international conference on 3d vision (3DV), pages 402–411. IEEE, 2017

    2017

  8. [16]

    Get3d: A generative model of high quality 3d textured shapes learned from images.Advances In Neural Information Processing Systems, 35:31841–31854, 2022

    Jun Gao, Tianchang Shen, Zian Wang, Wenzheng Chen, Kangxue Yin, Daiqing Li, Or Litany, Zan Gojcic, and Sanja Fidler. Get3d: A generative model of high quality 3d textured shapes learned from images.Advances In Neural Information Processing Systems, 35:31841–31854, 2022

    2022

  9. [17]

    Generative adversarial networks

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde- Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks. Communications of the ACM, 63(11):139–144, 2020

    2020

  10. [18]

    Accurate, large minibatch sgd: Training imagenet in 1 hour.arXiv preprint arXiv:1706.02677, 2017

    Priya Goyal, Piotr Dollár, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He. Accurate, large minibatch sgd: Training imagenet in 1 hour.arXiv preprint arXiv:1706.02677, 2017

    2017 arXiv

  11. [19]

    Convolutional-network models to predict wall-bounded turbulence from wall quantities.Journal of Fluid Mechanics, 928:A27, 2021

    Luca Guastoni, Alejandro Güemes, Andrea Ianiro, Stefano Discetti, Philipp Schlat- ter, Hossein Azizpour, and Ricardo Vinuesa. Convolutional-network models to predict wall-bounded turbulence from wall quantities.Journal of Fluid Mechanics, 928:A27, 2021

    2021

  12. [20]

    From coarse wall measurements to turbulent velocity fields through deep learning.Physics of fluids, 33(7), 2021

    Alejandro Güemes, Stefano Discetti, Andrea Ianiro, Beril Sirmacek, Hossein Azizpour, and Ricardo Vinuesa. From coarse wall measurements to turbulent velocity fields through deep learning.Physics of fluids, 33(7), 2021

    2021

  13. [21]

    Active control for drag reduction of turbulent channel flow based on convolutional neural networks.Physics of Fluids, 32(9), 2020

    Bing-Zheng Han and Wei-Xi Huang. Active control for drag reduction of turbulent channel flow based on convolutional neural networks.Physics of Fluids, 32(9), 2020

    2020

  14. [22]

    Gnot: A general neural operator transformer for operator learning

    Zhongkai Hao, Zhengyi Wang, Hang Su, Chengyang Ying, Yinpeng Dong, Song- ming Liu, Ze Cheng, Jian Song, and Jun Zhu. Gnot: A general neural operator transformer for operator learning. InInternational Conference on Machine Learn- ing, pages 12556–12569. PMLR, 2023

    2023

  15. [23]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  16. [24]

    Zero-shot text-guided object generation with dream fields

    Ajay Jain, Ben Mildenhall, Jonathan T Barron, Pieter Abbeel, and Ben Poole. Zero-shot text-guided object generation with dream fields. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 867–876, 2022

    2022

  17. [25]

    Near-wall turbulence.Physics of Fluids, 25(10), 2013

    Javier Jiménez. Near-wall turbulence.Physics of Fluids, 25(10), 2013. 19

    2013

  18. [26]

    3d gaussian splatting for real-time radiance field rendering.ACM Trans

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

    2023

  19. [27]

    Deep fluids: A generative network for parameterized fluid simulations

    Byungsoo Kim, Vinicius C Azevedo, Nils Thuerey, Theodore Kim, Markus Gross, and Barbara Solenthaler. Deep fluids: A generative network for parameterized fluid simulations. InComputer graphics forum, volume 38, pages 59–70. Wiley Online Library, 2019

    2019

  20. [28]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  21. [29]

    Auto-encoding variational bayes.arXiv preprint arXiv:1312.6114, 2013

    Diederik P Kingma and Max Welling. Auto-encoding variational bayes.arXiv preprint arXiv:1312.6114, 2013

    2013 arXiv

  22. [30]

    Neural operator: Learning maps between function spaces with applications to pdes.Journal of Machine Learning Research, 24(89):1–97, 2023

    Nikola Kovachki, Zongyi Li, Burigede Liu, Kamyar Azizzadenesheli, Kaushik Bhat- tacharya, Andrew Stuart, and Anima Anandkumar. Neural operator: Learning maps between function spaces with applications to pdes.Journal of Machine Learning Research, 24(89):1–97, 2023

    2023

  23. [31]

    Deep learning in fluid dynamics.Journal of Fluid Mechanics, 814: 1–4, 2017

    J Nathan Kutz. Deep learning in fluid dynamics.Journal of Fluid Mechanics, 814: 1–4, 2017

    2017

  24. [32]

    Deep learning.nature, 521 (7553):436–444, 2015

    Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning.nature, 521 (7553):436–444, 2015

    2015

  25. [33]

    Transformer for partial differential equations’ operator learning.arXiv preprint arXiv:2205.13671, 2022

    Zijie Li, Kazem Meidani, and Amir Barati Farimani. Transformer for partial differential equations’ operator learning.arXiv preprint arXiv:2205.13671, 2022

    2022 arXiv

  26. [34]

    Fourier neural operator for parametric partial differential equations.arXiv preprint arXiv:2010.08895, 2020

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differential equations.arXiv preprint arXiv:2010.08895, 2020

    2010 arXiv

  27. [35]

    From zero to turbulence: Generative modeling for 3d flow simulation

    Marten Lienen, David Lüdke, Jan Hansen-Palmus, and Stephan Günnemann. From zero to turbulence: Generative modeling for 3d flow simulation. InThe Twelfth International Conference on Learning Representations

  28. [36]

    Deepmetahandles: Learning deformation meta-handles of 3d meshes with biharmonic coordinates

    Minghua Liu, Minhyuk Sung, Radomir Mech, and Hao Su. Deepmetahandles: Learning deformation meta-handles of 3d meshes with biharmonic coordinates. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12–21, 2021

    2021

  29. [37]

    One-2-3-45++: Fast single image to 3d objects with consistent multi-view generation and 3d diffusion

    Minghua Liu, Ruoxi Shi, Linghao Chen, Zhuoyang Zhang, Chao Xu, Xinyue Wei, Hansheng Chen, Chong Zeng, Jiayuan Gu, and Hao Su. One-2-3-45++: Fast single image to 3d objects with consistent multi-view generation and 3d diffusion. In Proceedings of the IEEE/CVF Conference on Comp...

    2024

  30. [38]

    Zero-1-to-3: Zero-shot one image to 3d object

    Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tokmakov, Sergey Zakharov, and Carl Vondrick. Zero-1-to-3: Zero-shot one image to 3d object. InProceedings of the IEEE/CVF international conference on computer vision, pages 9298–9309, 2023. 20

    2023

  31. [39]

    Meshdiffusion: Score-based generative 3d mesh modeling.arXiv preprint arXiv:2303.08133, 2023

    Zhen Liu, Yao Feng, Michael J Black, Derek Nowrouzezahrai, Liam Paull, and Weiyang Liu. Meshdiffusion: Score-based generative 3d mesh modeling.arXiv preprint arXiv:2303.08133, 2023

    2023 arXiv

  32. [40]

    Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

    2017 arXiv

  33. [41]

    Supervised convolutional network for three-dimensional fluid data reconstruction from sectional flow fields with adaptive super-resolution assistance

    Mitsuaki Matsuo, Taichi Nakamura, Masaki Morimoto, Kai Fukami, and Koji Fukagata. Supervised convolutional network for three-dimensional fluid data reconstruction from sectional flow fields with adaptive super-resolution assistance. arXiv preprint arXiv:2103.09020, 2021

    2021 arXiv

  34. [42]

    Pc2: Projection- conditioned point cloud diffusion for single-image 3d reconstruction

    Luke Melas-Kyriazi, Christian Rupprecht, and Andrea Vedaldi. Pc2: Projection- conditioned point cloud diffusion for single-image 3d reconstruction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12923–12932, 2023

    2023

  35. [43]

    Occupancy networks: Learning 3d reconstruction in function space

    Lars Mescheder, Michael Oechsle, Michael Niemeyer, Sebastian Nowozin, and Andreas Geiger. Occupancy networks: Learning 3d reconstruction in function space. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4460–4470, 2019

    2019

  36. [44]

    Nerf: Representing scenes as neural radiance fields for view synthesis.Communications of the ACM, 65(1):99–106, 2021

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

    2021

  37. [45]

    Dit-3d: Exploring plain diffusion transformers for 3d shape generation

    Shentong Mo, Enze Xie, Ruihang Chu, Lanqing Hong, Matthias Niessner, and Zhenguo Li. Dit-3d: Exploring plain diffusion transformers for 3d shape generation. Advances in neural information processing systems, 36:67960–67971, 2023

    2023

  38. [46]

    Multidiff: Consistent novel view synthesis from a single image

    Norman Müller, Katja Schwarz, Barbara Rössle, Lorenzo Porzi, Samuel Rota Bulò, Matthias Nießner, and Peter Kontschieder. Multidiff: Consistent novel view synthesis from a single image. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1...

    2024

  39. [47]

    Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow.Physics of Fluids, 33(2), 2021

    Taichi Nakamura, Kai Fukami, Kazuto Hasegawa, Yusuke Nabae, and Koji Fuka- gata. Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow.Physics of Fluids, 33(2), 2021

    2021

  40. [48]

    Point-e: A system for generating 3d point clouds from complex prompts.arXiv preprint arXiv:2212.08751, 2022

    Alex Nichol, Heewoo Jun, Prafulla Dhariwal, Pamela Mishkin, and Mark Chen. Point-e: A system for generating 3d point clouds from complex prompts.arXiv preprint arXiv:2212.08751, 2022

    2022 arXiv

  41. [49]

    Radsplat: Radiance field-informed gaussian splatting for robust real-time rendering with 900+ fps.arXiv preprint arXiv:2403.13806, 2024

    Michael Niemeyer, Fabian Manhardt, Marie-Julie Rakotosaona, Michael Oechsle, Daniel Duckworth, Rama Gosula, Keisuke Tateno, John Bates, Dominik Kaeser, and Federico Tombari. Radsplat: Radiance field-informed gaussian splatting for robust real-time rendering with 900+ fps.arXiv...

    2024 arXiv

  42. [50]

    Cfdnet: A deep learning-based accelerator for fluid simulations

    Octavi Obiols-Sales, Abhinav Vishnu, Nicholas Malaya, and Aparna Chan- dramowliswharan. Cfdnet: A deep learning-based accelerator for fluid simulations. In Proceedings of the 34th ACM international conference on supercomputing, pages 1–12, 2020. 21

    2020

  43. [51]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 4195–4205, 2023

    2023

  44. [52]

    Turbulent flows.Measurement Science and Technology, 12(11): 2020–2021, 2001

    Stephen B Pope. Turbulent flows.Measurement Science and Technology, 12(11): 2020–2021, 2001

    2020

  45. [53]

    MaziarRaissi, ParisPerdikaris, andGeorgeEKarniadakis. Physics-informedneural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations.Journal of Computational physics, 378:686–707, 2019

    2019

  46. [54]

    Acceleration of a full-scale industrial cfd application with op2.IEEE Transactions on Parallel and Distributed Systems, 27 (5):1265–1278, 2015

    István Z Reguly, Gihan R Mudalige, Carlo Bertolli, Michael B Giles, Adam Betts, Paul HJ Kelly, and David Radford. Acceleration of a full-scale industrial cfd application with op2.IEEE Transactions on Parallel and Distributed Systems, 27 (5):1265–1278, 2015

    2015

  47. [55]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. InPro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  48. [56]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. InMedical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part I...

    2015

  49. [57]

    Clip-forge: Towards zero-shot text-to- shape generation

    Aditya Sanghi, Hang Chu, Joseph G Lambourne, Ye Wang, Chin-Yi Cheng, Marco Fumero, and Kamal Rahimi Malekshan. Clip-forge: Towards zero-shot text-to- shape generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18603–18613, 2022

    2022

  50. [58]

    Tomographic piv: principles and practice.Measurement Science and Technology, 24(1):012001, 2012

    Fulvio Scarano. Tomographic piv: principles and practice.Measurement Science and Technology, 24(1):012001, 2012

    2012

  51. [59]

    Shake-the-box: Lagrangian particle tracking at high particle image densities.Experiments in fluids, 57:1–27, 2016

    Daniel Schanz, Sebastian Gesemann, and Andreas Schröder. Shake-the-box: Lagrangian particle tracking at high particle image densities.Experiments in fluids, 57:1–27, 2016

    2016

  52. [60]

    Zero123++: a single image to consistent multi-view diffusion base model.arXiv preprint arXiv:2310.15110, 2023

    Ruoxi Shi, Hansheng Chen, Zhuoyang Zhang, Minghua Liu, Chao Xu, Xinyue Wei, Linghao Chen, Chong Zeng, and Hao Su. Zero123++: a single image to consistent multi-view diffusion base model.arXiv preprint arXiv:2310.15110, 2023

    2023 arXiv

  53. [61]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermodynamics. InInterna- tional conference on machine learning, pages 2256–2265. PMLR, 2015

    2015

  54. [62]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020. 22

    2011 arXiv

  55. [63]

    Learned coarse models for efficient turbulence simulation.arXiv e-prints, pages arXiv–2112, 2021

    Kimberly Stachenfeld, Drummond B Fielding, Dmitrii Kochkov, Miles Cranmer, Tobias Pfaff, Jonathan Godwin, Can Cui, Shirley Ho, Peter Battaglia, and Alvaro Sanchez-Gonzalez. Learned coarse models for efficient turbulence simulation.arXiv e-prints, pages arXiv–2112, 2021

    2021

  56. [64]

    Pdebench: An exten- sive benchmark for scientific machine learning.Advances in Neural Information Processing Systems, 35:1596–1611, 2022

    Makoto Takamoto, Timothy Praditia, Raphael Leiteritz, Daniel MacKinlay, Francesco Alesiani, Dirk Pflüger, and Mathias Niepert. Pdebench: An exten- sive benchmark for scientific machine learning.Advances in Neural Information Processing Systems, 35:1596–1611, 2022

    2022

  57. [65]

    Sa- convonet: Sign-agnostic optimization of convolutional occupancy networks

    Jiapeng Tang, Jiabao Lei, Dan Xu, Feiying Ma, Kui Jia, and Lei Zhang. Sa- convonet: Sign-agnostic optimization of convolutional occupancy networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6504–6513, 2021

    2021

  58. [66]

    Accelerating eulerian fluid simulation with convolutional networks

    Jonathan Tompson, Kristofer Schlachter, Pablo Sprechmann, and Ken Perlin. Accelerating eulerian fluid simulation with convolutional networks. InInternational conference on machine learning, pages 3424–3433. PMLR, 2017

    2017

  59. [67]

    Factorized fourier neural operators

    Alasdair Tran, Alexander Mathews, Lexing Xie, and Cheng Soon Ong. Factorized fourier neural operators. InThe Eleventh International Conference on Learning Representations

  60. [68]

    Patchnets: Patch-based generalizable deep implicit 3d shape representations

    Edgar Tretschk, Ayush Tewari, Vladislav Golyanik, Michael Zollhöfer, Carsten Stoll, and Christian Theobalt. Patchnets: Patch-based generalizable deep implicit 3d shape representations. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Pro...

    2020

  61. [69]

    Lion: Latent point diffusion models for 3d shape generation.Advances in Neural Information Processing Systems, 35:10021–10039, 2022

    Arash Vahdat, Francis Williams, Zan Gojcic, Or Litany, Sanja Fidler, Karsten Kreis, et al. Lion: Latent point diffusion models for 3d shape generation.Advances in Neural Information Processing Systems, 35:10021–10039, 2022

    2022

  62. [70]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017

    2017

  63. [71]

    Enhancing computational fluid dynamics with machine learning.Nature Computational Science, 2(6):358–366, 2022

    Ricardo Vinuesa and Steven L Brunton. Enhancing computational fluid dynamics with machine learning.Nature Computational Science, 2(6):358–366, 2022

    2022

  64. [72]

    The transformative potential of machine learning for experiments in fluid mechanics.Nature Reviews Physics, 5(9):536–545, 2023

    Ricardo Vinuesa, Steven L Brunton, and Beverley J McKeon. The transformative potential of machine learning for experiments in fluid mechanics.Nature Reviews Physics, 5(9):536–545, 2023

    2023

  65. [73]

    Sv3d: Novel multi-view synthesis and 3d generation from a single image using latent video diffusion, 2024

    Vikram Voleti, Chun-Han Yao, Mark Boss, Adam Letts, David Pankratz, Dmitry Tochilkin, Christian Laforte, Robin Rombach, and Varun Jampani. Sv3d: Novel multi-view synthesis and 3d generation from a single image using latent video diffusion, 2024. URL https://arxiv.org/abs/2403.12008

    2024 arXiv

  66. [74]

    Learning incompressible fluid dynamics from scratch-towards fast, differentiable fluid models that generalize

    Nils Wandel, Michael Weinmann, and Reinhard Klein. Learning incompressible fluid dynamics from scratch-towards fast, differentiable fluid models that generalize. In International Conference on Learning Representations. 23

  67. [75]

    Reduced-order modeling for turbulent wake of a finite wall-mounted square cylinder based on artificial neural network

    Mustafa Z Yousif and Hee-Chang Lim. Reduced-order modeling for turbulent wake of a finite wall-mounted square cylinder based on artificial neural network. Physics of Fluids, 34(1), 2022

    2022

  68. [76]

    A deep-learning approach for reconstructing 3d turbulent flows from 2d observation data

    Mustafa Z Yousif, Linqi Yu, Sergio Hoyas, Ricardo Vinuesa, and HeeChang Lim. A deep-learning approach for reconstructing 3d turbulent flows from 2d observation data. Scientific Reports, 13(1):2529, 2023

    2023

  69. [77]

    Mip-splatting: Alias-free 3d gaussian splatting

    Zehao Yu, Anpei Chen, Binbin Huang, Torsten Sattler, and Andreas Geiger. Mip-splatting: Alias-free 3d gaussian splatting. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 19447–19456, 2024

    2024

  70. [78]

    3dilg: Irregular latent grids for 3d generative modeling.Advances in Neural Information Processing Systems, 35: 21871–21885, 2022

    Biao Zhang, Matthias Nießner, and Peter Wonka. 3dilg: Irregular latent grids for 3d generative modeling.Advances in Neural Information Processing Systems, 35: 21871–21885, 2022

    2022

  71. [79]

    3dshape2vecset: A 3d shape representation for neural fields and generative diffusion models.ACM Transactions on Graphics (TOG), 42(4):1–16, 2023

    Biao Zhang, Jiapeng Tang, Matthias Niessner, and Peter Wonka. 3dshape2vecset: A 3d shape representation for neural fields and generative diffusion models.ACM Transactions on Graphics (TOG), 42(4):1–16, 2023. 24

    2023

Pith tools

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