pith. sign in

arxiv: 2606.20563 · v1 · pith:3PKSD7YHnew · submitted 2026-06-18 · 💻 cs.CV

JanusMesh: Fast and Zero-Shot 3D Visual Illusion Generation via Cross-Space Denoising

Siang-Ling Zhang , Huai-Hsun Cheng , Tsung-Ju Yang , Yu-Lun Liu This is my paper

Pith reviewed 2026-06-26 18:16 UTC · model grok-4.3

classification 💻 cs.CV
keywords 3D visual illusionstext-driven 3D generationdiffusion modelszero-shot generationmesh fusionSDF blendingCLIP guidanceview-conditioned synthesis
0
0 comments X

The pith

A cross-space dual-branch denoising process fuses two shapes into one coherent mesh then applies view-specific 2D textures to produce dual-semantic 3D illusions without training.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper aims to create single 3D meshes that reveal entirely different objects when viewed from different angles, driven by text prompts. It splits the task into a first stage that moves 3D latents into voxel space for CLIP-based alignment and SDF surface blending to join the shapes smoothly, then a second stage that projects multiple 2D diffusion outputs onto the joined geometry for consistent appearance. This targets the problems of slow optimization methods and visible seams in simple stitching approaches. If the stages work as described, users could obtain realistic illusions in a few minutes while preserving both geometric continuity and distinct semantics from each viewpoint.

Core claim

Our approach decouples the generation into two stages. First, we propose a cross-space dual-branch denoising process. This process dynamically decodes 3D latents into voxel space for CLIP-guided orientation alignment and Signed Distance Field (SDF) blending, which ensures seamless geometric fusion. Second, we introduce a view-conditioned texture synthesis module that projects and aggregates view-specific 2D diffusion priors onto the fused geometry. Extensive experiments demonstrate that our method generates highly realistic, dual-semantic 3D illusions in just 3-5 minutes and significantly outperforms existing methods in geometric integrity, semantic recognizability, and efficiency.

What carries the argument

Cross-space dual-branch denoising process that dynamically decodes 3D latents into voxel space for CLIP-guided orientation alignment and SDF blending, combined with a view-conditioned texture synthesis module that projects and aggregates 2D diffusion priors.

If this is right

  • Produces geometrically seamless fusion of two distinct shapes into one mesh
  • Maintains dual semantics through aggregation of view-specific 2D priors
  • Completes the full generation pipeline in 3-5 minutes
  • Achieves higher geometric integrity and semantic recognizability than optimization or stitching baselines

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The explicit separation of geometry fusion from texture synthesis could be reused for other multi-view consistency tasks in 3D content creation.
  • Because the method relies on existing 2D diffusion models for textures, its output quality tracks improvements in those models over time.
  • The dual-branch structure might be extended to support three or more distinct viewing angles by adding further alignment branches.

Load-bearing premise

Dynamically decoding 3D latents into voxel space and applying CLIP-guided alignment plus SDF blending will automatically produce geometrically coherent objects without seams or semantic leaks, and projecting 2D diffusion priors will preserve dual semantics on the fused mesh without additional training.

What would settle it

A generated mesh that shows visible seams at fusion boundaries or loses clear recognition of one intended semantic object when rendered from the corresponding viewpoint would show the method fails to deliver the claimed coherence.

Figures

Figures reproduced from arXiv: 2606.20563 by Huai-Hsun Cheng, Siang-Ling Zhang, Tsung-Ju Yang, Yu-Lun Liu.

Figure 1
Figure 1. Figure 1: Zero-shot 3D Visual Illusion Generation. [ [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Comparison of 3D visual illusion generation methods. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Pipeline overview. (a) Stage 1 employs dual-branch denoising. At each step, latents are decoded to voxel space, rotation-aligned, and fused via SDF blending ( [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: SDF blending. (a) Given two rotation-aligned voxels, we compute their SDFs, take the element-wise average, and binarize with threshold τ to obtain the blended voxel. (b) 2D illustration: averaging the SDFs of a circle (A) and a triangle (B) produces a smooth intermediate contour lying geometrically between the two. clean latent at each timestep t: \small x^1_{1|t} = z_t + u_\theta (z_t;\, t,\, y_1)(1 - t),… view at source ↗
Figure 5
Figure 5. Figure 5: Noise Guidance. Given two prompts y1 and y2, two single-semantic voxels vˆ1 and vˆ2 are independently generated and concatenated at θ1 and θ2 to form vguide, encoded into zguide via the Sparse Structure Encoder. Noise Blending Guidance mixes zguide with pure noise via α · zguide + (1 − α) · znoise, injecting a mild spatial prior. Space Control Guidance interpolates between zguide and noise at timestep t0, … view at source ↗
Figure 6
Figure 6. Figure 6: CLIP-guided Orientation Search. Anchor View Selection [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Extension to three-object 3D illusion. Our framework scales to three semantics (e.g., “Grapes”, “A Pineapple”, “Bamboo”) by adding a third denoising branch, with rotation angles fixed at 0 ◦ , 120◦ , 240◦ to uniformly cover 360◦ . Noise Guidance steers all three branches toward their respective targets, ensuring each semantic is clearly presented at its target viewpoint. angles at 0 ◦ , 120◦ , and 240◦ . A… view at source ↗
Figure 8
Figure 8. Figure 8: Object detection at the junction viewpoint for “Bamboo” & “Dwarf Cottage”. (a) Direct Con￾catenation is detected as two sepa￾rate objects (red and green boxes); (b) Ours is detected as a single uni￾fied object (red box only). (3) FID & KID. To measure visual real￾ism, we compute FID [22] and KID [2] be￾tween 1,000 renders of our results and 1,000 reference images (20 views of 50 objects) from Objaverse 1.0… view at source ↗
Figure 9
Figure 9. Figure 9: Qualitative comparison with baselines. Left-to-right: View 1, blended mesh, View 2. (a) Shape From Semantics [40] suffers from over-saturation and ge￾ometry leakage (e.g., “Stork”/“Grapes”). (b) Direct Concatenation exposes unnatural junction seams and the opposing geometry at target views. (c) Ours produces a single coherent mesh with clear, view-dependent semantics and no leakage. (d) TRELLIS [73], promp… view at source ↗
Figure 10
Figure 10. Figure 10: Qualitative comparison (CLIP-guided orientation). [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: User study. (Left) Semantic Recognizability [PITH_FULL_IMAGE:figures/full_fig_p014_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Ablation on geometry blending (“carriage”/“sundae”). Left-to-right: View 1, blended mesh, View 2. (a) Union: yields conflicting junctions. (b) Blur Avg: loses fine details. (c) Minkowski: over-expands geometry. (d) Polar Coord: distorts asymmetric objects. (e) SDF Avg (Ours): optimally balances geometric integrity and semantic preservation. “A Frog” & “A Turtle” Input (a) w/o noise guidance “Purple Succul… view at source ↗
Figure 13
Figure 13. Figure 13: Ablation on noise guidance across varying geometric compatibilities: (a) no guidance, (b) Noise Blending, and (c) Space Control. (c) handles large geo￾metric discrepancies (“Bamboo”/“Grapes”) best via strong spatial constraints. (b) is optimal for similar silhouettes but distinct semantics (“Succulent”/“Rooster”), avoiding semantic loss in (a) and residual artifacts in (c). For compatible pairs (“Frog”/“T… view at source ↗
Figure 14
Figure 14. Figure 14: Ablation on texture synthesis and orientation search. [PITH_FULL_IMAGE:figures/full_fig_p016_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Qualitative results of three-object 3D illusion generation. [PITH_FULL_IMAGE:figures/full_fig_p016_15.png] view at source ↗
read the original abstract

Creating 3D visual illusions, a single 3D mesh that reveals entirely different semantics from various viewing angles, is a fascinating but tough challenge. Existing optimization-based methods are slow and can produce oversaturated colors. In contrast, naive stitching approaches fail to produce geometrically coherent objects. This results in visible unnatural seams and semantic leaks. In this paper, we present a fast and training-free framework for generating text-driven 3D visual illusions. Our approach decouples the generation into two stages. First, we propose a cross-space dual-branch denoising process. This process dynamically decodes 3D latents into voxel space for CLIP-guided orientation alignment and Signed Distance Field (SDF) blending, which ensures seamless geometric fusion. Second, we introduce a view-conditioned texture synthesis module that projects and aggregates view-specific 2D diffusion priors onto the fused geometry. Extensive experiments demonstrate that our method generates highly realistic, dual-semantic 3D illusions in just 3-5 minutes. It significantly outperforms existing methods in geometric integrity, semantic recognizability, and efficiency. Project page: https://siang1105.github.io/JanusMesh.github.io/

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims to introduce JanusMesh, a fast training-free framework for text-driven 3D visual illusions. It decouples generation into (1) a cross-space dual-branch denoising process that dynamically decodes 3D latents into voxel space, applies CLIP-guided orientation alignment and SDF blending for seamless geometric fusion, and (2) a view-conditioned texture synthesis module that projects and aggregates 2D diffusion priors onto the fused mesh. The method is asserted to produce realistic dual-semantic 3D meshes in 3-5 minutes and to significantly outperform prior optimization-based and stitching methods on geometric integrity, semantic recognizability, and efficiency.

Significance. If the central claims hold, the work would be significant for providing an efficient zero-shot alternative to slow optimization pipelines and seam-prone stitching, with potential impact on AR/VR content creation and creative 3D design. The explicit decoupling of geometry fusion from texture synthesis is a clear conceptual contribution; the reported runtime is a practical strength.

major comments (2)
  1. [Abstract / §3] Abstract and §3 (method description): the cross-space dual-branch denoising process is presented only at the level of high-level stages (voxel decoding, CLIP alignment, SDF blending) with no equations, pseudocode, or explicit definition of the blending operator, latent initialization for the two text prompts, or the alignment loss. This mechanism is load-bearing for the central claim that the process 'ensures seamless geometric fusion' without post-processing or semantic leaks; its absence prevents verification of the contrast with naive stitching failures.
  2. [§4] §4 (experiments): the abstract states that 'extensive experiments demonstrate' significant outperformance and supplies runtime numbers, yet the provided text contains no tables of quantitative metrics (e.g., geometric error, CLIP similarity scores, user-study percentages), no ablation results, and no error bars or statistical tests. Without these, the performance claims cannot be assessed and remain unverified.
minor comments (2)
  1. [§3] Notation for the two text prompts and the dual-branch latents should be introduced explicitly at first use to avoid ambiguity.
  2. [Figures] Figure captions should include the exact text prompts used for each example to allow direct reproduction of the dual-semantic results.

Simulated Author's Rebuttal

2 responses · 0 unresolved

Thank you for the constructive feedback on our manuscript. We address each major comment below and will revise the paper to incorporate the requested details and evaluations.

read point-by-point responses

  1. Referee: [Abstract / §3] Abstract and §3 (method description): the cross-space dual-branch denoising process is presented only at the level of high-level stages (voxel decoding, CLIP alignment, SDF blending) with no equations, pseudocode, or explicit definition of the blending operator, latent initialization for the two text prompts, or the alignment loss. This mechanism is load-bearing for the central claim that the process 'ensures seamless geometric fusion' without post-processing or semantic leaks; its absence prevents verification of the contrast with naive stitching failures.

    Authors: We agree that the current description remains at a high level. In the revised manuscript we will add the explicit mathematical formulation of the cross-space dual-branch denoising, including latent initialization for the pair of text prompts, the CLIP-guided alignment loss, the precise definition of the SDF blending operator, and pseudocode for the overall procedure. These additions will make the mechanism verifiable and will clarify the distinction from naive stitching. revision: yes

  2. Referee: [§4] §4 (experiments): the abstract states that 'extensive experiments demonstrate' significant outperformance and supplies runtime numbers, yet the provided text contains no tables of quantitative metrics (e.g., geometric error, CLIP similarity scores, user-study percentages), no ablation results, and no error bars or statistical tests. Without these, the performance claims cannot be assessed and remain unverified.

    Authors: We acknowledge that the manuscript text as provided lacks quantitative tables, ablations, and statistical reporting. In the revision we will insert tables with geometric error metrics, CLIP-based semantic similarity scores, user-study percentages, component ablations, error bars, and statistical tests to substantiate the performance claims. revision: yes

Circularity Check

0 steps flagged

No circularity: method described as composition of existing components without reduction to fitted inputs or self-citation chains

full rationale

The provided abstract and description present a two-stage pipeline (cross-space dual-branch denoising with voxel decoding + CLIP alignment + SDF blending, followed by view-conditioned texture projection) as an empirical engineering combination of standard diffusion, CLIP, and SDF primitives. No equations, fitted parameters, or 'predictions' are shown that reduce by construction to the inputs. No self-citations or uniqueness theorems are invoked in the visible text. The central claim is therefore an assertion of effectiveness for the described composition, which can be evaluated against external benchmarks without internal circular reduction. This is the normal self-contained case.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Review performed on abstract only; no explicit free parameters, axioms, or invented entities are stated. The method implicitly relies on standard assumptions about CLIP and diffusion models.

axioms (2)
  • domain assumption CLIP embeddings can reliably guide orientation alignment between two semantic views inside voxel space
    Invoked in the first stage of the described denoising process.
  • domain assumption SDF blending of voxel-decoded latents produces seamless geometry without post-hoc fixes
    Claimed to ensure geometric fusion in the abstract.

pith-pipeline@v0.9.1-grok · 5752 in / 1510 out tokens · 29437 ms · 2026-06-26T18:16:39.183723+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

81 extracted references · 1 canonical work pages

  1. [1]

    Bar-Tal, O., Yariv, L., Lipman, Y., Dekel, T.: Multidiffusion: Fusing diffusion paths for controlled image generation (2023)

    2023

  2. [2]

    arXiv preprint arXiv:1801.01401 (2018)

    Bińkowski, M., Sutherland, D.J., Arbel, M., Gretton, A.: Demystifying mmd gans. arXiv preprint arXiv:1801.01401 (2018)

    Pith/arXiv arXiv 2018

  3. [3]

    In: ACM SIGGRAPH 2024 Conference Papers

    Burgert, R., Li, X., Leite, A., Ranasinghe, K., Ryoo, M.: Diffusion illusions: Hiding images in plain sight. In: ACM SIGGRAPH 2024 Conference Papers. pp. 1–11 (2024)

    2024

  4. [4]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Cao, T., Kreis, K., Fidler, S., Sharp, N., Yin, K.: Texfusion: Synthesizing 3d tex- tures with text-guided image diffusion models. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 4169–4181 (2023)

    2023

  5. [5]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Chang, P., Sancho, S., Tang, J., Gross, M., Azevedo, V.: Lookingglass: Generative anamorphoses via laplacian pyramid warping. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 24–33 (2025)

    2025

  6. [6]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Chen, D.Z., Siddiqui, Y., Lee, H.Y., Tulyakov, S., Nießner, M.: Text2tex: Text- driven texture synthesis via diffusion models. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 18558–18568 (2023)

    2023

  7. [7]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Chen, R., Chen, Y., Jiao, N., Jia, K.: Fantasia3d: Disentangling geometry and appearance for high-quality text-to-3d content creation. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 22246–22256 (2023)

    2023

  8. [8]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Chen, T.H., Chen, Y.H., Tu, T., Lee, J.Y., Wu, C.Y., Lin, F., Zhang, H., Paz, D., Huang, X., Guo, Y., et al.: Pantheon360: Taming digital twin generation via 3d-aware 360deg video diffusion. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11138–11149 (2026)

    2026

  9. [9]

    Advances in Neural Information Processing Systems37, 85045– 85073 (2024)

    Chen, Z., Geng, D., Owens, A.: Images that sound: Composing images and sounds on a single canvas. Advances in Neural Information Processing Systems37, 85045– 85073 (2024)

    2024

  10. [10]

    arXiv preprint arXiv:2602.12280 (2026)

    Cheng, H.H., Zhang, S.L., Liu, Y.L.: Stroke of surprise: Progressive semantic illu- sions in vector sketching. arXiv preprint arXiv:2602.12280 (2026)

    Pith/arXiv arXiv 2026

  11. [11]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Cheng, W., Mu, J., Zeng, X., Chen, X., Pang, A., Zhang, C., Wang, Z., Fu, B., Yu, G., Liu, Z., et al.: Mvpaint: Synchronized multi-view diffusion for painting anything 3d. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 585–594 (2025)

    2025

  12. [12]

    Debnath, S., Tiwari, A., Sadekar, K., Raman, S.: Rasp: revisiting 3d anamorphic artforshadow-guidedpackingofirregularobjects.In:ProceedingsoftheComputer Vision and Pattern Recognition Conference. pp. 5849–5858 (2025)

    2025

  13. [13]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Deitke, M., Schwenk, D., Salvador, J., Weihs, L., Michel, O., VanderBilt, E., Schmidt, L., Ehsani, K., Kembhavi, A., Farhadi, A.: Objaverse: A universe of annotated 3d objects. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13142–13153 (2023) 18 S.-L. Zhang et al

    2023

  14. [14]

    In: European conference on computer vision

    Deng, K., Omernick, T., Weiss, A., Ramanan, D., Zhu, J.Y., Zhou, T., Agrawala, M.: Flashtex: Fast relightable mesh texturing with lightcontrolnet. In: European conference on computer vision. pp. 90–107. Springer (2024)

    2024

  15. [15]

    ACM Transac- tions on Graphics (TOG)44(4), 1–10 (2025)

    Dodik, A., Yu, I., Chandra, K., Ragan-Kelley, J., Tenenbaum, J., Sitzmann, V., Solomon, J.: Meschers: Geometry processing of impossible objects. ACM Transac- tions on Graphics (TOG)44(4), 1–10 (2025)

    2025

  16. [16]

    In: International confer- ence on machine learning

    Du, Y., Durkan, C., Strudel, R., Tenenbaum, J.B., Dieleman, S., Fergus, R., Sohl- Dickstein, J., Doucet, A., Grathwohl, W.S.: Reduce, reuse, recycle: Compositional generation with energy-based diffusion models and mcmc. In: International confer- ence on machine learning. pp. 8489–8510. PMLR (2023)

    2023

  17. [17]

    arXiv preprint arXiv:2512.05343 (2025)

    Fedele, E., Engelmann, F., Huang, I., Litany, O., Pollefeys, M., Guibas, L.: Space- control: Introducing test-time spatial control to 3d generative modeling. arXiv preprint arXiv:2512.05343 (2025)

    arXiv 2025

  18. [18]

    arXiv preprint arXiv:2412.09625 (2024)

    Feng, Y., Sanjay, V., Lutz, S., AlBahar, B., Ge, S., Huang, J.B.: Illusion3d: 3d multiview illusion with 2d diffusion priors. arXiv preprint arXiv:2412.09625 (2024)

    arXiv 2024

  19. [19]

    In: Proceedings of the Com- puter Vision and Pattern Recognition Conference

    Gao, X., Yang, S., Liu, J.: Ptdiffusion: Free lunch for generating optical illusion hid- den pictures with phase-transferred diffusion model. In: Proceedings of the Com- puter Vision and Pattern Recognition Conference. pp. 18240–18249 (2025)

    2025

  20. [20]

    In: European Conference on Computer Vision

    Geng, D., Park, I., Owens, A.: Factorized diffusion: Perceptual illusions by noise de- composition. In: European Conference on Computer Vision. pp. 366–384. Springer (2024)

    2024

  21. [21]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Geng, D., Park, I., Owens, A.: Visual anagrams: Generating multi-view optical illusions with diffusion models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 24154–24163 (2024)

    2024

  22. [22]

    Advances in neural information processing systems30(2017)

    Heusel,M.,Ramsauer,H.,Unterthiner,T.,Nessler,B.,Hochreiter,S.:Ganstrained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems30(2017)

    2017

  23. [23]

    Advances in neural information processing systems33, 6840–6851 (2020)

    Ho, J., Jain, A., Abbeel, P.: Denoising diffusion probabilistic models. Advances in neural information processing systems33, 6840–6851 (2020)

    2020

  24. [24]

    arXiv preprint arXiv:2207.12598 (2022)

    Ho, J., Salimans, T.: Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598 (2022)

    Pith/arXiv arXiv 2022

  25. [25]

    arXiv preprint arXiv:2311.04400 (2023)

    Hong, Y., Zhang, K., Gu, J., Bi, S., Zhou, Y., Liu, D., Liu, F., Sunkavalli, K., Bui, T., Tan, H.: Lrm: Large reconstruction model for single image to 3d. arXiv preprint arXiv:2311.04400 (2023)

    Pith/arXiv arXiv 2023

  26. [26]

    ACM Trans

    Hsiao, K.W., Huang, J.B., Chu, H.K.: Multi-view wire art. ACM Trans. Graph. 37(6), 242 (2018)

    2018

  27. [27]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Huang, Y.C., Chan, J., Chien, H.J., Liu, Y.L.: Voxify3d: Pixel art meets volumetric rendering. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 15398–15410 (2026)

    2026

  28. [28]

    arXiv preprint arXiv:2512.25073 (2025)

    Huang, Y.C., Chien, H.J., Lin, C.Y., Chen, Y.H., Liu, Y.L.: Gamo: Geometry- aware multi-view diffusion outpainting for sparse-view 3d reconstruction. arXiv preprint arXiv:2512.25073 (2025)

    Pith/arXiv arXiv 2025

  29. [29]

    URLhttps://doi.org/10.5281/zenodo.5143773

    Ilharco, G., Wortsman, M., Wightman, R., Gordon, C., Carlini, N., Taori, R., Dave, A., Shankar, V., Namkoong, H., Miller, J., Hajishirzi, H., Farhadi, A., Schmidt, L.: Openclip (Jul 2021).https://doi.org/10.5281/zenodo.5143773,https://doi. org/10.5281/zenodo.5143773, if you use this software, please cite it as below

    work page doi:10.5281/zenodo.5143773 2021

  30. [30]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Jain, A., Mildenhall, B., Barron, J.T., Abbeel, P., Poole, B.: Zero-shot text-guided object generation with dream fields. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 867–876 (2022) JanusMesh 19

    2022

  31. [31]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Jain, A., Tancik, M., Abbeel, P.: Putting nerf on a diet: Semantically consistent few-shot view synthesis. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 5885–5894 (2021)

    2021

  32. [32]

    arXiv preprint arXiv:2305.02463 (2023)

    Jun, H., Nichol, A.: Shap-e: Generating conditional 3d implicit functions. arXiv preprint arXiv:2305.02463 (2023)

    Pith/arXiv arXiv 2023

  33. [33]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Ke, B.H., Xie, Y.Z., Liu, Y.L., Chiu, W.C.: Stealthattack: Robust 3d gaussian splatting poisoning via density-guided illusions. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 27400–27411 (2025)

    2025

  34. [34]

    arXiv preprint arXiv:2403.14370 (2024)

    Kim, J., Koo, J., Yeo, K., Sung, M.: Synctweedies: A general generative framework based on synchronized diffusions. arXiv preprint arXiv:2403.14370 (2024)

    arXiv 2024

  35. [35]

    arXiv preprint arXiv:2307.04787 (2023)

    Kim, S., Lee, K., Choi, J.S., Jeong, J., Sohn, K., Shin, J.: Collaborative score distillation for consistent visual synthesis. arXiv preprint arXiv:2307.04787 (2023)

    arXiv 2023

  36. [36]

    Lan, Y., Zhou, S., Lyu, Z., Hong, F., Yang, S., Dai, B., Pan, X., Loy, C.C.: Gaus- siananything: Interactive point cloud latent diffusion for 3d generation (2025)

    2025

  37. [37]

    arXiv preprint arXiv:2510.15869 (2025)

    Lee, J.Y., Liu, Y.R., Tsai, S.R., Chang, W.C., Wu, C.H., Chan, J., Zhao, Z., Lin, C.H., Liu, Y.L.: Skyfall-gs: Synthesizing immersive 3d urban scenes from satellite imagery. arXiv preprint arXiv:2510.15869 (2025)

    arXiv 2025

  38. [38]

    Advances in Neural Information Processing Systems36, 50648–50660 (2023)

    Lee, Y., Kim, K., Kim, H., Sung, M.: Syncdiffusion: Coherent montage via syn- chronized joint diffusions. Advances in Neural Information Processing Systems36, 50648–50660 (2023)

    2023

  39. [39]

    arXiv preprint arXiv:2502.00360 (2025)

    Li, L., Wang, C., Zhou, Y., Deng, B., Zhang, J.: Shape from semantics: 3d shape generation from multi-view semantics. arXiv preprint arXiv:2502.00360 (2025)

    arXiv 2025

  40. [40]

    Li, L., Wang, C., Zhou, Y., Deng, B., Zhang, J.: Shape from semantics: 3d shape generation from multi-view semantics (2025),https://arxiv.org/abs/2502. 00360

    2025

  41. [41]

    In: European Conference on Computer Vision

    Li, M.F., Ku, Y.F., Yen, H.X., Liu, C., Liu, Y.L., Chen, A.Y., Kuo, C.H., Sun, M.: Genrc: Generative 3d room completion from sparse image collections. In: European Conference on Computer Vision. pp. 146–163. Springer (2024)

    2024

  42. [42]

    In: 2025 International Conference on 3D Vision (3DV)

    Li, R., Han, J., Melas-Kyriazi, L., Sun, C., An, Z., Gui, Z., Sun, S., Torr, P., Jakab,T.:Dreambeast:Distilling3dfantasticalanimalswithpart-awareknowledge transfer. In: 2025 International Conference on 3D Vision (3DV). pp. 1243–1252. IEEE (2025)

    2025

  43. [43]

    IEEE Transactions on Pattern Analysis and Machine Intel- ligence (2025)

    Li, Y., Zou, Z.X., Liu, Z., Wang, D., Liang, Y., Yu, Z., Liu, X., Guo, Y.C., Liang, D., Ouyang, W., et al.: Triposg: High-fidelity 3d shape synthesis using large-scale rectified flow models. IEEE Transactions on Pattern Analysis and Machine Intel- ligence (2025)

    2025

  44. [44]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Liang, Y., Yang, X., Lin, J., Li, H., Xu, X., Chen, Y.: Luciddreamer: Towards high- fidelity text-to-3d generation via interval score matching. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 6517–6526 (2024)

    2024

  45. [45]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Lin, C.H., Gao, J., Tang, L., Takikawa, T., Zeng, X., Huang, X., Kreis, K., Fidler, S., Liu, M.Y., Lin, T.Y.: Magic3d: High-resolution text-to-3d content creation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 300–309 (2023)

    2023

  46. [46]

    arXiv preprint arXiv:2210.02747 (2022)

    Lipman, Y., Chen, R.T., Ben-Hamu, H., Nickel, M., Le, M.: Flow matching for generative modeling. arXiv preprint arXiv:2210.02747 (2022)

    Pith/arXiv arXiv 2022

  47. [47]

    In: European conference on computer vision

    Liu, N., Li, S., Du, Y., Torralba, A., Tenenbaum, J.B.: Compositional visual gen- eration with composable diffusion models. In: European conference on computer vision. pp. 423–439. Springer (2022) 20 S.-L. Zhang et al

    2022

  48. [48]

    In: Proceedings of the IEEE/CVF inter- national conference on computer vision

    Liu, R., Wu, R., Van Hoorick, B., Tokmakov, P., Zakharov, S., Vondrick, C.: Zero- 1-to-3: Zero-shot one image to 3d object. In: Proceedings of the IEEE/CVF inter- national conference on computer vision. pp. 9298–9309 (2023)

    2023

  49. [49]

    arXiv preprint arXiv:2209.03003 (2022)

    Liu, X., Gong, C., Liu, Q.: Flow straight and fast: Learning to generate and transfer data with rectified flow. arXiv preprint arXiv:2209.03003 (2022)

    Pith/arXiv arXiv 2022

  50. [50]

    arXiv preprint arXiv:2309.03453 (2023)

    Liu, Y., Lin, C., Zeng, Z., Long, X., Liu, L., Komura, T., Wang, W.: Syncdreamer: Generating multiview-consistent images from a single-view image. arXiv preprint arXiv:2309.03453 (2023)

    Pith/arXiv arXiv 2023

  51. [51]

    In: SIGGRAPH Asia 2024 Conference Papers

    Liu,Y.,Xie,M.,Liu,H.,Wong,T.T.:Text-guidedtexturingbysynchronizedmulti- view diffusion. In: SIGGRAPH Asia 2024 Conference Papers. pp. 1–11 (2024)

    2024

  52. [52]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Long, X., Guo, Y.C., Lin, C., Liu, Y., Dou, Z., Liu, L., Ma, Y., Zhang, S.H., Habermann, M., Theobalt, C., et al.: Wonder3d: Single image to 3d using cross- domain diffusion. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 9970–9980 (2024)

    2024

  53. [53]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Metzer, G., Richardson, E., Patashnik, O., Giryes, R., Cohen-Or, D.: Latent-nerf for shape-guided generation of 3d shapes and textures. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 12663– 12673 (2023)

    2023

  54. [54]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Michel, O., Bar-On, R., Liu, R., Benaim, S., Hanocka, R.: Text2mesh: Text-driven neural stylization for meshes. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13492–13502 (2022)

    2022

  55. [55]

    Advances in Neural Information Processing Systems36, 72983–73007 (2023)

    Minderer, M., Gritsenko, A., Houlsby, N.: Scaling open-vocabulary object detec- tion. Advances in Neural Information Processing Systems36, 72983–73007 (2023)

    2023

  56. [56]

    ACM Transactions on Graphics28(5), 156–1 (2009)

    Mitra, N.J., Pauly, M.: Shadow art. ACM Transactions on Graphics28(5), 156–1 (2009)

    2009

  57. [57]

    ACM Transactions on Graph- ics (TOG)25(3), 527–532 (2006)

    Oliva, A., Torralba, A., Schyns, P.G.: Hybrid images. ACM Transactions on Graph- ics (TOG)25(3), 527–532 (2006)

    2006

  58. [58]

    In: ACM SIGGRAPH 2023 conference proceedings

    Perroni-Scharf, M., Rusinkiewicz, S.: Constructing printable surfaces with view- dependent appearance. In: ACM SIGGRAPH 2023 conference proceedings. pp. 1–10 (2023)

    2023

  59. [59]

    arXiv preprint arXiv:2209.14988 (2022)

    Poole, B., Jain, A., Barron, J.T., Mildenhall, B.: Dreamfusion: Text-to-3d using 2d diffusion. arXiv preprint arXiv:2209.14988 (2022)

    Pith/arXiv arXiv 2022

  60. [60]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Qu, Z., Yang, L., Zhang, H., Xiang, T., Pang, K., Song, Y.Z.: Wired perspectives: Multi-view wire art embraces generative ai. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6149–6158 (2024)

    2024

  61. [61]

    In: International conference on machine learning

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al.: Learning transferable visual models from natural language supervision. In: International conference on machine learning. pp. 8748–8763. PmLR (2021)

    2021

  62. [62]

    In: ACM SIGGRAPH 2023 conference proceedings

    Richardson, E., Metzer, G., Alaluf, Y., Giryes, R., Cohen-Or, D.: Texture: Text- guided texturing of 3d shapes. In: ACM SIGGRAPH 2023 conference proceedings. pp. 1–11 (2023)

    2023

  63. [63]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

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

    2022

  64. [64]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Sadekar, K., Tiwari, A., Raman, S.: Shadow art revisited: a differentiable ren- dering based approach. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 29–37 (2022)

    2022

  65. [65]

    arXiv preprint arXiv:2308.16512 (2023) JanusMesh 21

    Shi,Y.,Wang,P.,Ye,J.,Long,M.,Li,K.,Yang,X.:Mvdream:Multi-viewdiffusion for 3d generation. arXiv preprint arXiv:2308.16512 (2023) JanusMesh 21

    Pith/arXiv arXiv 2023

  66. [66]

    arXiv preprint arXiv:2010.02502 (2020)

    Song, J., Meng, C., Ermon, S.: Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502 (2020)

    Pith/arXiv arXiv 2010

  67. [67]

    arXiv preprint arXiv:2309.16653 (2023)

    Tang, J., Ren, J., Zhou, H., Liu, Z., Zeng, G.: Dreamgaussian: Generative gaussian splatting for efficient 3d content creation. arXiv preprint arXiv:2309.16653 (2023)

    Pith/arXiv arXiv 2023

  68. [68]

    In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition

    Wang, C., Chai, M., He, M., Chen, D., Liao, J.: Clip-nerf: Text-and-image driven manipulation of neural radiance fields. In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition. pp. 3835–3844 (2022)

    2022

  69. [69]

    arXiv preprint arXiv:2411.19161 (2024)

    Wang, C., Deng, B., Zhang, J.: Neural shadow art. arXiv preprint arXiv:2411.19161 (2024)

    arXiv 2024

  70. [70]

    Advances in neural information processing systems36, 8406–8441 (2023)

    Wang, Z., Lu, C., Wang, Y., Bao, F., Li, C., Su, H., Zhu, J.: Prolificdreamer: High-fidelity and diverse text-to-3d generation with variational score distillation. Advances in neural information processing systems36, 8406–8441 (2023)

    2023

  71. [71]

    In: European conference on computer vision

    Wang, Z., Wang, Y., Chen, Y., Xiang, C., Chen, S., Yu, D., Li, C., Su, H., Zhu, J.: Crm: Single image to 3d textured mesh with convolutional reconstruction model. In: European conference on computer vision. pp. 57–74. Springer (2024)

    2024

  72. [72]

    Advances in Neural Information Processing Systems37, 121859–121881 (2024)

    Wu, S., Lin, Y., Zhang, F., Zeng, Y., Xu, J., Torr, P., Cao, X., Yao, Y.: Direct3d: Scalable image-to-3d generation via 3d latent diffusion transformer. Advances in Neural Information Processing Systems37, 121859–121881 (2024)

    2024

  73. [73]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Xiang, J., Lv, Z., Xu, S., Deng, Y., Wang, R., Zhang, B., Chen, D., Tong, X., Yang, J.: Structured 3d latents for scalable and versatile 3d generation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 21469–21480 (2025)

    2025

  74. [74]

    arXiv preprint arXiv:2404.07191 (2024)

    Xu, J., Cheng, W., Gao, Y., Wang, X., Gao, S., Shan, Y.: Instantmesh: Efficient 3d mesh generation from a single image with sparse-view large reconstruction models. arXiv preprint arXiv:2404.07191 (2024)

    Pith/arXiv arXiv 2024

  75. [75]

    arXiv preprint arXiv:2406.03293 (2024)

    Yang, X., Chen, C., Yang, X., Liu, F., Lin, G.: Text-to-image rectified flow as plug-and-play priors. arXiv preprint arXiv:2406.03293 (2024)

    arXiv 2024

  76. [76]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Yeh, Y.Y., Huang, J.B., Kim, C., Xiao, L., Nguyen-Phuoc, T., Khan, N., Zhang, C., Chandraker, M., Marshall, C.S., Dong, Z., et al.: Texturedreamer: Image- guided texture synthesis through geometry-aware diffusion. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4304– 4314 (2024)

    2024

  77. [77]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Yi, T., Fang, J., Wang, J., Wu, G., Xie, L., Zhang, X., Liu, W., Tian, Q., Wang, X.: Gaussiandreamer: Fast generation from text to 3d gaussians by bridging 2d and 3d diffusion models. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 6796–6807 (2024)

    2024

  78. [78]

    ACM Transactions on Graphics (TOG)43(6), 1–14 (2024)

    Yu, X., Yuan, Z., Guo, Y.C., Liu, Y.T., Liu, J., Li, Y., Cao, Y.P., Liang, D., Qi, X.: Texgen: a generative diffusion model for mesh textures. ACM Transactions on Graphics (TOG)43(6), 1–14 (2024)

    2024

  79. [79]

    In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition

    Zeng, X., Chen, X., Qi, Z., Liu, W., Zhao, Z., Wang, Z., Fu, B., Liu, Y., Yu, G.: Paint3d: Paint anything 3d with lighting-less texture diffusion models. In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 4252–4262 (2024)

    2024

  80. [80]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Zhang, L., Rao, A., Agrawala, M.: Adding conditional control to text-to-image diffusion models. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 3836–3847 (2023)

    2023

Showing first 80 references.