Pith. sign in

REVIEW 3 major objections 6 minor 5 cited by

PhyCAGE: Physically Plausible Compositional 3D Asset Generation from a Single Image

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read PhyCAGE claims to be the first single-image method that generates compositional 3D assets as separate Gaussian Splatting models corrected by physics simulation, eliminating object penetration.

desk verdict PhyCAGE's core algorithm is described with a sign error that makes it gradient ascent as written, and the physical plausibility claim is never quantitatively supported; the direction is promising but the paper needs major revision. read the letter →

arxiv 2411.18548 v1 pith:H3AADNP2 submitted 2024-11-27 cs.CV

Han Yan , Mingrui Zhang , Yang Li , Chao Ma , Pan Ji This is my paper · ORCID
classification cs.CV
keywords compositional3Dgenerationsingle-imagereconstructionGaussiansplattingscoredistillationsamplingmaterialpointmethodphysicalplausibilityobjectpenetrationavoidancemulti-viewdiffusion
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

PhyCAGE is a pipeline that takes a single photograph of two interacting objects and produces two separate 3D Gaussian Splatting models whose relative positions are physically plausible. The paper's claim is that this is the first single-image method to generate compositional 3D assets without object penetration, and that the key to this result is a new optimization rule called PSE-SDS (Physical Simulation-Enhanced Score Distillation Sampling). Instead of moving Gaussian centers directly along image-loss gradients, the gradient is fed into a Material Point Method simulator as the initial velocity, so the simulator, not the optimizer, decides where each particle settles. If correct, this makes editable, simulation-ready 3D assets obtainable from ordinary images, which matters for VR, gaming, and robotics applications that need separate objects with stable contact.

What carries the argument

The central mechanism is PSE-SDS (Physical Simulation-Enhanced Score Distillation Sampling), an optimization rule that delegates Gaussian position updates to physics. At each optimization step k, the current centers are set as the simulation's initial positions x0, the gradient of the total loss with respect to the centers is set as the initial velocity v0, and N substeps of the Material Point Method advance the particles; the final positions become the new Gaussian centers. The first MPM substep is mathematically just one gradient-descent step with step size Δt = γ/N, and the remaining substeps are what convert a visual pressure into a physically consistent rearrangement by solving momentum and contact constraints. Appearance parameters (opacity and color) are frozen, and only the scale and rotation parameters receive an ordinary gradient update, so the simulator changes where objects are without repainting them.

What would settle it

Run the method on a set of single images with severe foreground-background occlusion, then measure the volume of Gaussian overlap after optimization: if many outputs still interpenetrate, whether inside the MPM simulator or after transfer to a different physics engine, the central claim fails. A complementary test is to sweep the undocumented material parameters; if the penetration-free property holds only in a narrow stiffness band, the method's physical plausibility is tied to an unstated tuning choice rather than to the PSE-SDS rule itself.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that physically plausible composition can be imposed after reconstruction rather than being baked into a generative model. The authors first segment the foreground object in the input image, inpaint the occluded background, and generate multi-view images for both components; these are fitted as two 3D Gaussian Splatting representations, with the foreground Gaussian centers initialized from a semantic signed-distance field. They then run PSE-SDS, where the gradient of the combined image and SDS loss with respect to Gaussian centers becomes the initial velocity of an MLS-MPM simulation, and the simulation output replaces the position update. The paper reports quantitative superiority over the two baselines on PSNR and CLIP score, and qualitative demonstrations that the generated objects do not interpenetrate when simulated. The contribution is framed as the first single-image compositional 3D generation method that avoids penetration.

Load-bearing premise

The load-bearing premise is that a short Material Point Method simulation, started with the image-loss gradient as its initial push and run with material stiffness parameters the paper never reports, will settle into a state that is both free of object interpenetration and still matches the input image.

Editorial extensions

If this is right

  • A single image of two interacting objects can yield two separate, independently editable 3D assets that remain stable under gravity and contact rather than a single fused mesh.
  • Because the correction happens through non-rigid simulation, the method can handle spatial relationships that a rigid similarity transform cannot, including deformation and support contact.
  • The same PSE-SDS rule can be applied iteratively, per the paper, to extend from two objects to larger compositions by treating one object as foreground and the rest as background in each pass.
  • The method does not require ground-truth 3D supervision; it relies only on the input image, multi-view diffusion, and a physical simulator, so it inherits improvements in any of those components.

Reading between the lines

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

  • An implication the paper leaves implicit is that the gradient-as-initial-velocity trick is a general way to fuse any differentiable visual loss with any differentiable simulator, so the same design could make text-only or video-based generators physics-aware by swapping the simulator while keeping PSE-SDS unchanged.
  • Because the paper leaves material parameters unspecified, stiffness, damping, and mass act as hidden hyperparameters that control how aggressively the simulator overrides the image loss; a natural extension is to estimate these parameters per object automatically from contact cues in the input image.
  • The paper demonstrates physical plausibility inside the same MPM simulator used for correction; a stronger external test would be to export the produced assets into an unrelated rigid-body or finite-element simulator and check whether the contact remains penetration-free.
  • Since the final quality is bounded by the multi-view generator's consistency, the method's ceiling is effectively the quality of that generator, and pairing PSE-SDS with a stronger reconstruction model, as the paper suggests, should directly improve the penetration-free rate.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper presents PhyCAGE, a method for generating two compositional 3D Gaussian Splatting assets from a single image while enforcing physical plausibility. The pipeline first segments the foreground object, inpaints the occluded background, generates multi-view images, and fits two separate 3DGS models. A new optimization scheme, Physical Simulation-Enhanced SDS (PSE-SDS), is introduced to update Gaussian positions by using the gradient of an SDS-plus-image loss as the initial velocity of an MPM simulation. The authors claim that this is the first method to produce compositional 3D assets without interpenetration from a single image. Experiments report higher PSNR and CLIP scores than Part123 and ComboVerse, and ablations compare PPPS, vanilla SDS, PPPS+SDS, and the proposed PSE-SDS.

Significance. If the central algorithmic description were correct, this would be a promising direction: the combination of 3DGS, SDS, and MPM is natural for interactive assets, and the paper makes a clear case that purely visual supervision leads to penetration. The ablation design is instructive, separating post-processing, visual-only, alternating, and coupled optimization. However, the current manuscript lacks support for its core claim. The described update rule appears to be gradient ascent, no quantitative physical metric is provided, and the MPM material configuration is undocumented. These are not cosmetic issues; they affect whether the method, as written, can produce the reported results. The paper does not ship code or machine-checked proof, so the burden of clarity is higher.

major comments (3)
  1. [Algorithm 1, Sec. 4.3] The core update rule is described with a wrong sign for gradient descent. Line 4 sets v0 = ∇θkµ L, and Eq. (9) updates positions as x^{t+1} = x^t + Δt v^{t+1}; hence the first substep moves the Gaussians by +Δt∇µL, i.e., in the direction of steepest ascent of L. Minimizing L in Eq. (19) requires moving opposite the gradient. The statement that this is 'equivalent to one step of vanilla optimization using gradient descent' is therefore incorrect. If implemented literally, PSE-SDS would push the asset away from the SDS and image supervision, which is incompatible with the high PSNR values in Table 1. Please correct the sign (e.g., v0 = −∇θkµ L) or clearly state the convention used, and confirm which version was implemented.
  2. [Sec. 5.2, Tables 1-2] The paper's central contribution is physical plausibility, explicitly 'without penetration' (Contributions, Sec. 1), but no quantitative physical metric is reported. Table 1 and Table 2 contain only PSNR and CLIP scores, which measure image similarity, not contact, penetration, or stability. The qualitative figures show red circles marking penetrations, but such hand-marked visual evidence does not substantiate 'no penetration.' Please add metrics such as interpenetration volume, penetration depth, number of penetrating Gaussian centers, contact area, or stability under further gravity simulation, and report them for the baselines and ablations.
  3. [Sec. 4.3, Eq. (20); Sec. 5.1] The MPM simulation is not specified sufficiently for reproducibility or for assessing physical plausibility. Eq. (20) leaves ψ undefined as 'particle mass, particle volume and materials parameters,' and the implementation details in Sec. 5.1 do not give values. The SDF boundary constraint referenced in Sec. 4.2 is also not described in the optimization. Without the material model (e.g., elastic constants, plastic parameters), the particle masses/volumes, the MPM grid resolution, and the number of substeps N, the simulator's behavior is unspecified and the physical claims cannot be verified. Please provide the full simulation configuration or a precise pointer to a reference implementation.
minor comments (6)
  1. [Sec. 4.2, Eq. (16)] The notation 'µ ∈ V2' is confusing; it seems to say the centers of G2 are initialized from the vertex set V2, but µ is also used for the centers. Please rephrase, e.g., 'with centers initialized from V2.'
  2. [Sec. 2.2] There is a typo in the related work: 'a the SMPL body pror works' should likely be 'prior works.' Please proofread.
  3. [Sec. 5.1] The guidance scale for the inpainting model is given as a set {7.5, 8.0, 9.0, 12.5}; it is unclear whether these are values per example or a sweep. Please clarify.
  4. [Algorithm 1, Sec. 4.3] The values of the learning rate γ and the number of MPM substeps N, which appear in Algorithm 1, are not reported in Sec. 5.1. Please provide the exact values used in the experiments.
  5. [Fig. 5] The caption refers to penetration marked by a red circle, but the figure as printed does not clearly show it; please increase visual clarity or add zoomed insets.
  6. [Sec. 5.2, Table 2] The CLIP-text and CLIP-ip metrics use the same text prompt and inpainted image that were used for SDS optimization, so those numbers partially measure self-consistency with the generation condition. Please acknowledge this dependency or report scores with an independent prompt set.

Circularity Check

2 steps flagged · score 4.0 of 10

Core PSE-SDS optimization is not circular, but two headline evaluation metrics (PSNR and CLIPip) compare rendered outputs against images that were used to fit the scored representations, making the evaluation partially self-referential.

  1. fitted input called prediction [Sec. 4.3 (Eq. 18) and Sec. 5.2 (PSNR metric)]
    "LImage = (1 − λ1)L1(I c, I) + λ1LSSIM (I c, I), where I c is the image rendered from the generated composed object, I denotes the original input image (Eq. 18); "Peak Signal-to-Noise Ratio (PSNR), which quantifies the similarity between the rendered image and the input image at the reference view" (Sec. 5.2)."

    The PSNR reported in Table 1 (30.70 dB) is computed on the reference view whose rendering is directly supervised by LImage in Eq. (18). Because the optimization explicitly minimizes L1 and SSIM between the rendered composed object and the original input image at that same view, the PSNR value is essentially a measure of training fit rather than an independent test of generalization or physical plausibility. The score is therefore forced upward by construction and does not by itself validate the physics-guided optimizer.

  2. fitted input called prediction [Sec. 4.1-4.2 (Eqs. 11 and 15) and Sec. 5.2 (CLIPip metric)]
    "G1 = GaussianSplatting({ ˆIi}16_i=1) (Eq. 15); "CLIPip... between the reference view of O1 and the inpainted image" (Sec. 5.2)."

    The inpainted images {ˆIi} are produced from the same input image and the same prompt τ1 used for the object (Eq. 11), and those exact images are used to fit G1 in Eq. (15). The CLIPip metric then compares the reference view rendered from G1 against those same inpainted images. A high CLIPip score therefore only confirms that G1 closely reproduces its own training views; it does not provide independent evidence that PSE-SDS generates a novel, physically plausible compositional asset.

full rationale

The core PhyCAGE pipeline is not circular: given an input image, it generates multi-view images with SyncDreamer, fits 3D Gaussian Splatting representations, and then optimizes Gaussian positions using an objective (Eq. 19) that combines image supervision and Score Distillation Sampling, with updates delegated to an MPM simulation. This is an optimization against external supervision, not a derivation that reduces to its own inputs. No load-bearing claim is justified solely by a self-citation; the cited works by the same authors are limited to related-work context and are not used to force the method's choices. The noted sign inconsistency in Algorithm 1 (setting v0 to the positive loss gradient and then advancing positions with a forward MPM step) is a correctness concern rather than a circularity, because it does not make the output equivalent to the input by construction. The genuine circularity is confined to the evaluation: PSNR is computed on the exact reference view minimized by LImage, and CLIPip compares renders of G1 against the inpainted images that were used to fit G1. These metrics are partly self-referential and should be interpreted as reconstruction fidelity measures rather than independent predictions. The paper's central claim of penetration-free, physically plausible compositional generation still has independent qualitative and ablation-based support, so the overall circularity score is moderate rather than high.

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

The central claim relies on several unstated modeling choices, most importantly the MPM material parameters and the optimization schedule. These are free parameters rather than derived quantities, and they materially affect whether the output is physically plausible. In addition, the pipeline depends on the reliability of several external components (SyncDreamer, Grounded-SAM, Part123), which are domain assumptions rather than proven facts.

free parameters (4)
  • lambda_1, lambda_2, lambda_3 = 0.2, 1.0, 0.00001
    Loss weights in Eq. (1), (2), (19) chosen empirically without sensitivity analysis.
  • SDS guidance scale and timestep range = guidance scale in {7.5, 8.0, 9.0, 12.5}, timestep decreases from 100 to 20
    Chosen by hand for the inpainting and SDS stages; no ablation on these values.
  • MPM material parameters psi
    Particle mass, volume, elastic modulus, etc. are not specified anywhere in the paper, yet they determine the simulated physical behavior.
  • Learning rate gamma and simulation substeps N
    gamma and N in Algorithm 1 control the substep time delta_t; their values are not reported.
assumptions (5)
  • domain assumption SyncDreamer generates multi-view images that are geometrically consistent with the input image and with each other.
    Used in Sec. 4.1, Eqs. (12)-(13) as the sole source of 3D reconstruction; the paper itself notes that output quality depends on this generator.
  • domain assumption Grounded-SAM segmentation is accurate enough to separate foreground from background and to provide consistent semantic maps across views.
    Used in Sec. 4.1, Eq. (10); inconsistent segmentation would corrupt the NeuS semantics and the subsequent GS fitting.
  • domain assumption Part123's NeuS reconstruction yields a 3D-consistent semantic field with accurate mesh vertices.
    Used in Sec. 4.2, Eq. (14); the GS centers of G2 are fixed to these vertices, so errors propagate directly into the final asset.
  • domain assumption The MPM simulation with Gaussians as particles and with assumed material parameters is a faithful model of real-world contact and deformation.
    Used in Sec. 4.3, Eq. (20); the material model and parameters are not specified, so the claimed physical plausibility is not well-defined.
  • domain assumption The coupled SDS-gradient-plus-simulation update converges to a fixed point that minimizes a blend of visual and physical objectives.
    Algorithm 1 assumes the simulation acts as a benign optimizer; no convergence analysis or stability guarantee is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PhyCAGE: Physically Plausible Compositional 3D Asset Generation from a Single Image." pith.science (2026). https://pith.science/paper/H3AADNP2

@misc{pith2026241118548,
  author       = {Pith},
  title        = {Pith review of: PhyCAGE: Physically Plausible Compositional 3D Asset Generation from a Single Image},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H3AADNP2}},
  note         = {Machine review of arXiv:2411.18548}
}
read the original abstract

We present PhyCAGE, the first approach for physically plausible compositional 3D asset generation from a single image. Given an input image, we first generate consistent multi-view images for components of the assets. These images are then fitted with 3D Gaussian Splatting representations. To ensure that the Gaussians representing objects are physically compatible with each other, we introduce a Physical Simulation-Enhanced Score Distillation Sampling (PSE-SDS) technique to further optimize the positions of the Gaussians. It is achieved by setting the gradient of the SDS loss as the initial velocity of the physical simulation, allowing the simulator to act as a physics-guided optimizer that progressively corrects the Gaussians' positions to a physically compatible state. Experimental results demonstrate that the proposed method can generate physically plausible compositional 3D assets given a single image.

Figures

Figures reproduced from arXiv: 2411.18548 by the authors.

Figure 1
Figure 1. PhyCAGE can generate compositional 3D assets with interactive objects in a physically plausible manner. The generated 3D [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overview of PhyCAGE. Given an input image, we first generate consistent multi-view images for the components of the assets (see Sec. 4.1). Then, we fit multi-view images with 3D Gaussian Splatting representations (see Sec. 4.2). Finally, we introduce a Physical Simulation-Enhanced SDS to further optimize the positions of the Gaussians (see Sec. 4.3). where g denotes the gravity acceleration. The updated ve￾locit… view at source ↗
Figure 3
Figure 3. Object inpainting with image diffusion models [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The overview of our PSE-SDS. The gradients come from the SDS and image loss are divided into two streams during the backpropagation. Specifically, ∇θk µ L is utilized as the initial velocity of the physical simulation for updating the positions µ of Gaussians. Image Lo…
Figure 5
Figure 5. Figure 5: Qualitative comparison with previous work. The green box illustrates the decomposed objects, while the blue box highlights the physical relationships, such as whether the components are in penetration (in red circle). Since we use 3DGS representation, we convert Gaussi…
Figure 6
Figure 6. Figure 6: Various methods of integrating interactive informa￾tion through physical simulation [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Ablation studies on LImage and LSDS [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Qualitative results of ablation studies on PSE-SDS. 6. Conclusion In this paper, we present PhyCAGE, the first approach to generate physically plausible compositional 3D assets from [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: More results. The blue boxes depict the input image, whereas the orange boxes showcase the generated outcomes fea￾turing decomposed objects. The colored point clouds provide vi￾sualizations of our generated 3DGS. a single Image. Our method incorporates a novel Physical…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

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

  1. PartDiffuser: Part-wise 3D Mesh Generation via Discrete Diffusion

    cs.CV 2025-11 unverdicted novelty 7.0 of 10

    PartDiffuser is a semi-autoregressive discrete diffusion framework that generates high-fidelity 3D meshes from point clouds by combining inter-part autoregression with intra-part parallel diffusion using a part-aware ...

  2. BANG: Dividing 3D Assets via Generative Exploded Dynamics

    cs.GR 2025-07 conditional novelty 7.0 of 10

    A diffusion-based method that generates smooth exploded-view sequences of 3D objects, enabling part-level decomposition, control, and reassembly.

  3. PhysX-CoT: Structured Physical Reasoning from a Single Image to Simulation-Ready 3D Assets

    cs.RO 2026-08 conditional novelty 6.0 of 10

    PhysX-CoT turns single-image 3D asset generation into an explicit, ordered, supervised chain of physical states, beating an output-centric VLM baseline on geometry and physical attributes.

  4. Muses: Designing, Composing, Generating Nonexistent Fantasy 3D Creatures without Training

    cs.CV 2026-01 conditional novelty 6.0 of 10

    Muses creates new fantasy 3D animals by designing a combined skeleton, fusing voxel parts from separate 3D models along that skeleton, then restyling textures via image editing — with no training.

  5. Generative Physical AI in Vision: A Survey

    cs.CV 2025-01 conditional novelty 6.0 of 10

    A structured review that categorizes physics-aware generative models in vision into explicit-simulation and implicit-learning families and proposes six integration paradigms.

Reference graph

Works this paper leans on

55 extracted references · 21 canonical work pages · cited by 5 Pith papers

  1. [1]

    Nonlinear Continuum Mechanics for Finite Element Analysis

    Javier Bonet and Richard D Wood. Nonlinear Continuum Mechanics for Finite Element Analysis. Cambridge Univer- sity Press, 1997. 3

    work page 1997

  2. [2]

    Comboverse: Compositional 3d assets creation using spatially-aware diffusion guidance

    Yongwei Chen, Tengfei Wang, Tong Wu, Xingang Pan, Kui Jia, and Ziwei Liu. Comboverse: Compositional 3d assets creation using spatially-aware diffusion guidance. arXiv:2403.12409, 2024. 2, 6

    arXiv 2024

  3. [3]

    Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication

    Yunuo Chen, Tianyi Xie, Zeshun Zong, Xuan Li, Feng Gao, Yin Yang, Ying Nian Wu, and Chenfanfu Jiang. Atlas3d: Physically constrained self-supporting text-to-3d for simula- tion and fabrication. arXiv:2405.18515, 2024. 3

    work page Pith review arXiv 2024

  4. [4]

    Set-the-scene: Global-local training for generating controllable nerf scenes

    Dana Cohen-Bar, Elad Richardson, Gal Metzer, Raja Giryes, and Daniel Cohen-Or. Set-the-scene: Global-local training for generating controllable nerf scenes. In ICCV, 2023. 2

    work page 2023

  5. [5]

    Tela: Text to layer-wise 3d clothed human generation

    Junting Dong, Qi Fang, Zehuan Huang, Xudong Xu, Jingbo Wang, Sida Peng, and Bo Dai. Tela: Text to layer-wise 3d clothed human generation. arXiv:2404.16748, 2024. 2

    arXiv 2024

  6. [6]

    Disentangled 3d scene generation with layout learning

    Dave Epstein, Ben Poole, Ben Mildenhall, Alexei A Efros, and Aleksander Holynski. Disentangled 3d scene generation with layout learning. arXiv:2402.16936, 2024. 2

    arXiv 2024

  7. [7]

    Silly rubber: an implicit material point method for simulating non- equilibrated viscoelastic and elastoplastic solids.ACM TOG, 38(4):1–13, 2019

    Yu Fang, Minchen Li, Ming Gao, and Chenfanfu Jiang. Silly rubber: an implicit material point method for simulating non- equilibrated viscoelastic and elastoplastic solids.ACM TOG, 38(4):1–13, 2019. 3

    work page 2019

  8. [8]

    Learning disentangled avatars with hybrid 3d representations

    Yao Feng, Weiyang Liu, Timo Bolkart, Jinlong Yang, Marc Pollefeys, and Michael J Black. Learning disentangled avatars with hybrid 3d representations. arXiv:2309.06441,

    arXiv

Show all 55 references
  1. [9]

    Dis- tance fields for rapid collision detection in physically based modeling

    Arnulph Fuhrmann, Gerrit Sobotka, and Clemens Groß. Dis- tance fields for rapid collision detection in physically based modeling. In Proceedings of GraphiCon, pages 58–65, 2003. 5

    2003

  2. [10]

    Graphdreamer: Compositional 3d scene synthesis from scene graphs

    Gege Gao, Weiyang Liu, Anpei Chen, Andreas Geiger, and Bernhard Sch ¨olkopf. Graphdreamer: Compositional 3d scene synthesis from scene graphs. In CVPR, 2024. 2

    2024

  3. [11]

    Physi- cally compatible 3d object modeling from a single image

    Minghao Guo, Bohan Wang, Pingchuan Ma, Tianyuan Zhang, Crystal Elaine Owens, Chuang Gan, Joshua B Tenenbaum, Kaiming He, and Wojciech Matusik. Physi- cally compatible 3d object modeling from a single image. arXiv:2405.20510, 2024. 3

    2024 arXiv

  4. [12]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. In NeurIPS, 2020. 1, 2

    2020

  5. [13]

    Lrm: Large reconstruction model for single image to 3d

    Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou, Difan Liu, Feng Liu, Kalyan Sunkavalli, Trung Bui, and Hao Tan. Lrm: Large reconstruction model for single image to 3d. arXiv:2311.04400, 2023. 1, 2

    2023 arXiv

  6. [14]

    Human- liff: Layer-wise 3d human generation with diffusion model

    Shoukang Hu, Fangzhou Hong, Tao Hu, Liang Pan, Haiyi Mei, Weiye Xiao, Lei Yang, and Ziwei Liu. Human- liff: Layer-wise 3d human generation with diffusion model. arXiv:2308.09712, 2023. 2

    2023 arXiv

  7. [15]

    A moving least squares material point method with displacement disconti- nuity and two-way rigid body coupling

    Yuanming Hu, Yu Fang, Ziheng Ge, Ziyin Qu, Yixin Zhu, Andre Pradhana, and Chenfanfu Jiang. A moving least squares material point method with displacement disconti- nuity and two-way rigid body coupling. ACM TOG, 37(4): 1–14, 2018. 3, 6

    2018

  8. [16]

    Jiang, C

    C. Jiang, C. Schroeder, A. Selle, J. Teran, and A. Stomakhin. The affine particle-in-cell method. ACM TOG, 34(4):1–10, 2015

    2015

  9. [17]

    Jiang, T

    C. Jiang, T. Gast, and J. Teran. Anisotropic elastoplasticity for cloth, knit and hair frictional contact. ACM TOG, 36(4): 1–14, 2017. 3

    2017

  10. [18]

    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 TOG, 42(4), 2023. 2, 3

    2023

  11. [19]

    Berg, Wan-Yen Lo, Piotr Doll ´ar, and Ross Girshick

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C. Berg, Wan-Yen Lo, Piotr Doll ´ar, and Ross Girshick. Segment anything. arXiv:2304.02643, 2023. 4

    2023 arXiv

  12. [20]

    Minchen Li, Zachary Ferguson, Teseo Schneider, Timothy Langlois, Denis Zorin, Daniele Panozzo, Chenfanfu Jiang, and Danny M. Kaufman. Incremental potential contact: Intersection- and inversion-free large deformation dynamics. ACM TOG, 39(4), 2020. 8

    2020

  13. [21]

    Part123: Part-aware 3d reconstruction from a single-view image

    Anran Liu, Cheng Lin, Yuan Liu, Xiaoxiao Long, Zhiyang Dou, Hao-Xiang Guo, Ping Luo, and Wenping Wang. Part123: Part-aware 3d reconstruction from a single-view image. In ACM SIGGRAPH, pages 1–12, 2024. 1, 5, 6

    2024

  14. [22]

    Physics3d: Learn- ing physical properties of 3d gaussians via video diffusion

    Fangfu Liu, Hanyang Wang, Shunyu Yao, Shengjun Zhang, Jie Zhou, and Yueqi Duan. Physics3d: Learn- ing physical properties of 3d gaussians via video diffusion. arXiv:2406.04338, 2024. 3

    2024 arXiv

  15. [23]

    Grounding dino: Marrying dino with grounded pre-training for open-set object detection

    Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, et al. Grounding dino: Marrying dino with grounded pre-training for open-set object detection. arXiv:2303.05499, 2023. 4

    2023 arXiv

  16. [24]

    Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age

    Yuan Liu, Cheng Lin, Zijiao Zeng, Xiaoxiao Long, Lingjie Liu, Taku Komura, and Wenping Wang. Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age. arXiv:2309.03453, 2023. 1, 2, 4

    2023 arXiv

  17. [25]

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J. Black. SMPL: A skinned multi- person linear model. SIGGRAPH Asia, 2015. 2

    2015

  18. [26]

    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

  19. [27]

    Phyrecon: Physically plausible neural scene reconstruction

    Junfeng Ni, Yixin Chen, Bohan Jing, Nan Jiang, Bin Wang, Bo Dai, Yixin Zhu, Song-Chun Zhu, and Siyuan Huang. Phyrecon: Physically plausible neural scene reconstruction. arXiv:2404.16666, 2024. 3

    2024 arXiv

  20. [28]

    Compositional 3d scene generation using locally conditioned diffusion

    Ryan Po and Gordon Wetzstein. Compositional 3d scene generation using locally conditioned diffusion. arXiv:2303.12218, 2023. 2

    2023 arXiv

  21. [29]

    Dreamfusion: Text-to-3d using 2d diffusion

    Ben Poole, Ajay Jain, Jonathan T Barron, and Ben Mildenhall. Dreamfusion: Text-to-3d using 2d diffusion. arXiv:2209.14988, 2022. 2, 5

    2022 arXiv

  22. [30]

    Magic123: One image to high-quality 3d object generation using both 2d and 3d diffusion priors

    Guocheng Qian, Jinjie Mai, Abdullah Hamdi, Jian Ren, Aliaksandr Siarohin, Bing Li, Hsin-Ying Lee, Ivan Sko- rokhodov, Peter Wonka, Sergey Tulyakov, et al. Magic123: One image to high-quality 3d object generation using both 2d and 3d diffusion priors. arXiv:2306.17843, 2023. 2

    2023 arXiv

  23. [31]

    Feature splatting: Language-driven physics-based scene syn- thesis and editing

    Ri-Zhao Qiu, Ge Yang, Weijia Zeng, and Xiaolong Wang. Feature splatting: Language-driven physics-based scene syn- thesis and editing. arXiv:2404.01223, 2024. 3

    2024 arXiv

  24. [32]

    Learning trans- ferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sas- try, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning trans- ferable visual models from natural language supervision. arXiv:2103.00020...

    2021 arXiv

  25. [33]

    D. Ram, T. Gast, C. Jiang, C. Schroeder, A. Stomakhin, J. Teran, and P. Kavehpour. A material point method for vis- coelastic fluids, foams and sponges. In SCA, 2015. 3

    2015

  26. [34]

    Grounded sam: Assembling open-world models for diverse visual tasks

    Tianhe Ren, Shilong Liu, Ailing Zeng, Jing Lin, Kun- chang Li, He Cao, Jiayu Chen, Xinyu Huang, Yukang Chen, Feng Yan, Zhaoyang Zeng, Hao Zhang, Feng Li, Jie Yang, Hongyang Li, Qing Jiang, and Lei Zhang. Grounded sam: Assembling open-world models for diverse visual tasks. arXi...

    2024 arXiv

  27. [35]

    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, 4

    2022

  28. [36]

    Zero123++: a single image to consistent multi-view dif- fusion base model

    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 dif- fusion base model. arXiv:2310.15110, 2023. 1, 2

    2023 arXiv

  29. [37]

    Fem simulation of 3d deformable solids: a practitioner’s guide to theory, dis- cretization and model reduction

    Eftychios Sifakis and Jernej Barbic. Fem simulation of 3d deformable solids: a practitioner’s guide to theory, dis- cretization and model reduction. In ACM SIGGRAPH 2012 Courses, pages 1–50, 2012. 8

    2012

  30. [38]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermodynamics. In ICML, 2015. 2

    2015

  31. [39]

    Dreamgaussian: Generative gaussian splatting for ef- ficient 3d content creation

    Jiaxiang Tang, Jiawei Ren, Hang Zhou, Ziwei Liu, and Gang Zeng. Dreamgaussian: Generative gaussian splatting for ef- ficient 3d content creation. arXiv:2309.16653, 2023. 1

    2023 arXiv

  32. [40]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In NeurIPS, 2017. 1

    2017

  33. [41]

    Disentangled clothed avatar generation from text de- scriptions

    Jionghao Wang, Yuan Liu, Zhiyang Dou, Zhengming Yu, Yongqing Liang, Xin Li, Wenping Wang, Rong Xie, and Li Song. Disentangled clothed avatar generation from text de- scriptions. arXiv:2312.05295, 2023. 2

    2023 arXiv

  34. [42]

    Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction

    Peng Wang, Lingjie Liu, Yuan Liu, Christian Theobalt, Taku Komura, and Wenping Wang. Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction. arXiv:2106.10689, 2021. 5

    2021 arXiv

  35. [43]

    Rodin: A generative model for sculpting 3d digital avatars using diffusion

    Tengfei Wang, Bo Zhang, Ting Zhang, Shuyang Gu, Jianmin Bao, Tadas Baltrusaitis, Jingjing Shen, Dong Chen, Fang Wen, Qifeng Chen, et al. Rodin: A generative model for sculpting 3d digital avatars using diffusion. In CVPR, 2023. 1, 2

    2023

  36. [44]

    Humancoser: Layered 3d human generation via semantic-aware diffusion model

    Yi Wang, Jian Ma, Ruizhi Shao, Qiao Feng, Yu-Kun Lai, Yebin Liu, and Kun Li. Humancoser: Layered 3d human generation via semantic-aware diffusion model. arXiv:2312.05804, 2023. 2

    2023 arXiv

  37. [45]

    Object- compositional neural implicit surfaces

    Qianyi Wu, Xian Liu, Yuedong Chen, Kejie Li, Chuanxia Zheng, Jianfei Cai, and Jianmin Zheng. Object- compositional neural implicit surfaces. In ECCV, 2022. 2

    2022

  38. [46]

    Objectsdf++: Improved object-compositional neural implicit surfaces

    Qianyi Wu, Kaisiyuan Wang, Kejie Li, Jianmin Zheng, and Jianfei Cai. Objectsdf++: Improved object-compositional neural implicit surfaces. In ICCV, 2023. 2

    2023

  39. [47]

    Assetfield: Assets mining and recon- figuration in ground feature plane representation

    Yuanbo Xiangli, Linning Xu, Xingang Pan, Nanxuan Zhao, Bo Dai, and Dahua Lin. Assetfield: Assets mining and recon- figuration in ground feature plane representation. In ICCV,

  40. [48]

    Physgaussian: Physics- integrated 3d gaussians for generative dynamics

    Tianyi Xie, Zeshun Zong, Yuxing Qiu, Xuan Li, Yutao Feng, Yin Yang, and Chenfanfu Jiang. Physgaussian: Physics- integrated 3d gaussians for generative dynamics. In CVPR,

  41. [49]

    Frankenstein: Generating semantic- compositional 3d scenes in one tri-plane

    Han Yan, Yang Li, Zhennan Wu, Shenzhou Chen, Weix- uan Sun, Taizhang Shang, Weizhe Liu, Tian Chen, Xiaqiang Dai, Chao Ma, et al. Frankenstein: Generating semantic- compositional 3d scenes in one tri-plane. arXiv:2403.16210,

    arXiv

  42. [50]

    Lam, Yan-Pei Cao, and Xi- hui Liu

    Yunhan Yang, Yukun Huang, Yuan-Chen Guo, Liangjun Lu, Xiaoyang Wu, Edmund Y . Lam, Yan-Pei Cao, and Xi- hui Liu. Sampart3d: Segment any part in 3d objects. arXiv:2411.07184, 2024. 1

    2024 arXiv

  43. [51]

    3dshape2vecset: A 3d shape representation for neu- ral fields and generative diffusion models

    Biao Zhang, Jiapeng Tang, Matthias Niessner, and Peter Wonka. 3dshape2vecset: A 3d shape representation for neu- ral fields and generative diffusion models. ACM TOG, 42 (4):1–16, 2023. 2

    2023

  44. [52]

    Clay: A controllable large-scale generative model for creating high-quality 3d assets

    Longwen Zhang, Ziyu Wang, Qixuan Zhang, Qiwei Qiu, Anqi Pang, Haoran Jiang, Wei Yang, Lan Xu, and Jingyi Yu. Clay: A controllable large-scale generative model for creating high-quality 3d assets. arXiv:2406.13897, 2024. 2

    2024 arXiv

  45. [53]

    Scenewiz3d: Towards text-guided 3d scene composition

    Qihang Zhang, Chaoyang Wang, Aliaksandr Siarohin, Peiye Zhuang, Yinghao Xu, Ceyuan Yang, Dahua Lin, Bolei Zhou, Sergey Tulyakov, and Hsin-Ying Lee. Scenewiz3d: Towards text-guided 3d scene composition. arXiv:2312.08885, 2023. 2

    2023 arXiv

  46. [54]

    Physdreamer: Physics-based interac- tion with 3d objects via video generation

    Tianyuan Zhang, Hong-Xing Yu, Rundi Wu, Brandon Y Feng, Changxi Zheng, Noah Snavely, Jiajun Wu, and William T Freeman. Physdreamer: Physics-based interac- tion with 3d objects via video generation. In ECCV, 2025. 3

    2025

  47. [55]

    Locally atten- tional sdf diffusion for controllable 3d shape generation

    Xin-Yang Zheng, Hao Pan, Peng-Shuai Wang, Xin Tong, Yang Liu, and Heung-Yeung Shum. Locally atten- tional sdf diffusion for controllable 3d shape generation. arXiv:2305.04461, 2023. 2

    2023 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.