pith. sign in

arxiv: 2606.01910 · v1 · pith:Q43RW57Wnew · submitted 2026-06-01 · 💻 cs.GR · cs.CV

Single-Line Drawing Generation via Semantics-Driven Optimization

Tanguy Magne , Alexandre Binninger , Ruben Wiersma , Olga Sorkine-Hornung This is my paper

Pith reviewed 2026-06-28 11:52 UTC · model grok-4.3

classification 💻 cs.GR cs.CV
keywords single-line drawingsvector graphicsscore distillation samplingURBS curvestext-to-image generationcontinuous line artsemantics-driven optimizationcurve optimization
0
0 comments X

The pith

Score distillation sampling on uniform rational B-spline parameters generates single-line vector drawings from text prompts or images.

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

The paper seeks to establish that optimizing the control points and weights of a uniform rational B-spline curve through score distillation sampling can automatically produce a single continuous stroke whose shape matches the semantics of a guiding text prompt or image. A sympathetic reader would care because line drawings distill complex subjects into minimal expressive marks, yet manual creation is time-intensive and existing automated methods often yield disconnected strokes or raster outputs that require extra conversion steps. If the claim holds, the resulting vector curves become immediately usable in physical production pipelines without manual cleanup. The method further incorporates style-guiding loss terms and claims measurable gains in visual appeal and fidelity to human continuous-line artists over current baselines.

Core claim

By applying score distillation sampling directly to the parameters of a uniform rational B-spline curve, the approach produces single-line drawings in vector format that follow either a text description or an input image; the representation guarantees one continuous stroke by construction, supplies editable control over detail level, and yields results rated higher in aesthetics and stylistic match to continuous-line artists than those from text-to-image diffusion models or prior optimization pipelines.

What carries the argument

Score distillation sampling applied to uniform rational B-spline (URBS) curve parameters, which enforces a single continuous stroke by design while incorporating semantic guidance and additional losses for artistic style.

If this is right

  • The output is a vector curve that can be fed directly into fabrication processes such as embroidery, laser engraving, and wire bending.
  • The spline representation supplies explicit, fine-grained control over the amount of detail retained in the drawing.
  • Additional loss terms can be added to steer the final artistic style without changing the single-stroke constraint.
  • Guidance works equally from a text prompt describing the concept or from an input image depicting it.

Where Pith is reading between the lines

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

  • Because the output remains a compact parametric curve, downstream editing tools could allow users to adjust individual segments while preserving the single-stroke property.
  • The same optimization loop might be applied to other parametric representations if the URBS form proves limiting for certain subjects.
  • Vector output opens the possibility of consistent line art across multiple viewpoints of a 3D model without re-rendering.

Load-bearing premise

Score distillation sampling applied to URBS curve parameters will reliably produce drawings that are both more aesthetically pleasing and more faithful to continuous-line artist style than existing text-to-image or optimization baselines.

What would settle it

A side-by-side comparison in which participants consistently rate outputs from standard text-to-image diffusion models as more aesthetically pleasing or closer to continuous-line artist style than the URBS-optimized curves would falsify the superiority claim.

Figures

Figures reproduced from arXiv: 2606.01910 by Alexandre Binninger, Olga Sorkine-Hornung, Ruben Wiersma, Tanguy Magne.

Figure 1
Figure 1. Figure 1: Single-line drawings, where lifting the pen from the paper is not allowed, lead to minimalist creations, often exhibiting surprising and expressive abstractions. Our method generates single-line drawings in vector format based on text and/or image input. From left to right: Input image and prompt, the single-line drawing at various stages of the optimization, the final result (the parameterization of the c… view at source ↗
Figure 3
Figure 3. Figure 3: Left: Gemini generates drawings with multiple curves (highlighted by red circles) in raster format. Middle: Vectorizing the raster output with Adobe Illustrator produces many curves (illus￾trated with different colors). Right: Connecting these curves gives a messy single-line drawing, with unnecessary strokes [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Left: ControlSketch [AFCO∗ 25] generates drawings with multiple curves. Middle left: Connecting these curves gives a messy single-line drawing, full of overdrawn and unnecessary strokes. Middle right: Directly generating single-line drawing with Control￾Sketch produces poor results. Right: Our result. style. The single Bézier curve does not fit single-line drawings well and connecting multiple Bézier strok… view at source ↗
Figure 5
Figure 5. Figure 5 [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Two uniform rational B-spline curves. Left: the control points in blue have very low associated weights, leading to straight lines and sharp corners near them. Right: removing these low￾weight control points results in a smoother-looking curve. The basis function is given as Bi,0(t) = ( 1 if ti ≤ t < ti+1, 0 otherwise. Bi,k(t) = t −ti ti+k −ti Bi,k−1(t) + ti+k+1 −t ti+k+1 −ti+1 Bi+1,k−1(t). In our case, we… view at source ↗
Figure 7
Figure 7. Figure 7: Left: an image generated with the original weights of the diffusion model. Right: a raster image generated using the LoRA weights trained on a dataset of single-line drawings. Both were generated with the same prompt: “A single line drawing of a group of mushrooms of different sizes growing together, depicted with one flowing line.” et al. [TSBU24] used this repulsion to enforce the fabricability of their … view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative comparison of various line drawing generation methods. Text-to-image models (Gemini in this case) produce raster images and contain multiple lines. We present here the results obtained after applying the single-line conversion strategy (see Sec. 4) [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Visual ablation study. Removing any component from our pipeline leads to worse-looking curves and too many details. connection strategy to the output of these text-to-image models, their image-image similarity decreases a bit and their visual scores increase a little, which is expected. However the visual quality metrics still remain similar or worse than ours. 4.3. Perceptual study To verify the quality o… view at source ↗
Figure 11
Figure 11. Figure 11: Different initialization strategies lead to very different results. Top row: initial curve. Bottom row: final optimized curve. input ours w/o repulsion loss [PITH_FULL_IMAGE:figures/full_fig_p009_11.png] view at source ↗
Figure 10
Figure 10. Figure 10: Different curve representations. these ablations in [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
Figure 15
Figure 15. Figure 15: Our method can be modified to obtain curves with vary￾ing width along the line [PITH_FULL_IMAGE:figures/full_fig_p010_15.png] view at source ↗
Figure 14
Figure 14. Figure 14: Our method can generate single-line drawings with horizontal start and end (top), allowing to connect them easily or to repeat a motif (bottom). where disabling the repulsion loss changes the style of the generated single-line drawing. Without the repulsion loss, we observe more cusps and fewer loops. This gives the petal of the rose and the feather of the flamingo a more appealing and realistic look [PI… view at source ↗
Figure 18
Figure 18. Figure 18: , our method produces vectorized curves that can be directly interpreted by digital embroidery machines. Furthermore, the flex￾ibility of our representation allows different stitching styles (e.g., satin or bean stitches) to be applied along the same path, resulting in a variety of visual textures while preserving the one-line con￾straint. The same benefit applies to wood laser engraving, where maintainin… view at source ↗
Figure 19
Figure 19. Figure 19: For niche subjects, such as this salamander, our method inherits the flaws of the diffusion model and struggles to create a good-looking result with meaningful inside details. For reference, a single drawing created by an artist is shown on the right. 4.9. Limitations and future work Our method is able to generate single-line drawings that follow the aesthetic of single-line drawings created by artists an… view at source ↗
Figure 20
Figure 20. Figure 20: Qualitative comparison of various line drawing generation methods. Text-to-image models (Gemini in this case) produce raster images and contain multiple lines. We present here the results obtained after applying the single-line conversion strategy (see Sec. 4). © 2026 The Author(s). Computer Graphics Forum published by Eurographics and John Wiley & Sons Ltd [PITH_FULL_IMAGE:figures/full_fig_p013_20.png] view at source ↗
Figure 21
Figure 21. Figure 21: Sample from the training data of the LoRA model. Top: target image. Bottom: input prompt. time step t. The gradient of the SDS loss is then given by ∇θLSDS = Et,ϵ  [PITH_FULL_IMAGE:figures/full_fig_p017_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: Left: an image generated with the original weights of the diffusion model. Middle: an image generated using the LoRA weights trained on a dataset of single-line drawings. Right: the result of the LoRA model after applying line vectorization and the line connection strategy. The diffusion model used for computing the SDS loss is Stable Diffusion 3.5 (SD3.5) [EKB∗ 24]. On a machine with a 32-core Intel i9-1… view at source ↗
Figure 23
Figure 23. Figure 23: , top). The second part contains 15 questions and forms the core of the study. Here, participants are presented with 4 images, each created with a different method, as described in Sec. 4.3. All of these images are actual single-line drawings (in the sense that the vector representation contains a single stroke). Participants have to rank them based on their adherence to the single-line drawing aesthetic … view at source ↗
read the original abstract

Line drawings are a highly expressive art form that requires the artist to abstract and distill the essence of their subject. We present the first semantics-driven method for automatically generating single-line drawings in vector format, guided either by a text prompt describing the concept or an input image depicting it. Our approach leverages score distillation sampling to optimize the parameters of a uniform rational B-spline (URBS) curve, ensuring that the drawing consists of a single continuous stroke by design. This representation provides fine-grained control over the level of detail, while additional loss terms allow us to steer the final artistic style. We demonstrate that our method outperforms state-of-the-art text-to-image models and optimization pipelines for this task, producing results that are both more aesthetically pleasing and more faithful to the style of continuous line drawing artists. Furthermore, because our method generates a vectorized curve, it directly supports downstream fabrication processes such as embroidery, laser engraving and wire bending. Our code and results are available at https://github.com/tanguymagne/SLDgen.

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 presents the first semantics-driven method for generating single-line drawings in vector format. It optimizes the parameters of a uniform rational B-spline (URBS) curve via score distillation sampling (SDS) from either a text prompt or input image, enforcing a single continuous stroke by construction while using additional loss terms to control artistic style. The method is claimed to outperform text-to-image models and optimization baselines in aesthetics and fidelity to continuous-line artist style, with direct support for fabrication tasks such as embroidery and laser engraving; code is released.

Significance. If the results hold, the contribution would be notable for generative graphics and computational art by introducing a vector parametric approach to single-line drawings that is semantics-driven and fabrication-ready. The by-construction continuity and code release are strengths that support reproducibility and downstream use.

major comments (2)
  1. [Abstract / Results] Abstract and results sections: the central claim that the method 'outperforms state-of-the-art text-to-image models and optimization pipelines' in producing 'more aesthetically pleasing' and 'more faithful' drawings is load-bearing but appears supported only by qualitative examples; no quantitative metrics, user studies, or ablation tables are referenced to substantiate superiority over baselines.
  2. [Method] Method description: while SDS on URBS parameters is a reasonable extension, the manuscript does not detail how the URBS control points and weights are initialized or regularized to avoid degenerate single-stroke solutions (e.g., self-intersections or collapsed segments), which could undermine the 'by design' continuity claim in practice.
minor comments (2)
  1. [Abstract / Method] The acronym 'URBS' is used throughout; confirm whether this is intended as a variant of NURBS or a typo for 'NURBS' (uniform rational B-splines).
  2. [Figures] Figure captions and comparison panels should explicitly label the input prompt/image and the competing methods for each example to improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive feedback. We address the two major comments below and will revise the manuscript to incorporate the suggested improvements.

read point-by-point responses

  1. Referee: [Abstract / Results] Abstract and results sections: the central claim that the method 'outperforms state-of-the-art text-to-image models and optimization pipelines' in producing 'more aesthetically pleasing' and 'more faithful' drawings is load-bearing but appears supported only by qualitative examples; no quantitative metrics, user studies, or ablation tables are referenced to substantiate superiority over baselines.

    Authors: We agree that the superiority claims would benefit from quantitative backing beyond the qualitative examples currently presented. The manuscript demonstrates advantages through visual comparisons, but to strengthen this load-bearing claim we will add a user study evaluating aesthetic preference and fidelity to continuous-line artist style, plus ablation tables on the loss terms, in the revised results section. revision: yes

  2. Referee: [Method] Method description: while SDS on URBS parameters is a reasonable extension, the manuscript does not detail how the URBS control points and weights are initialized or regularized to avoid degenerate single-stroke solutions (e.g., self-intersections or collapsed segments), which could undermine the 'by design' continuity claim in practice.

    Authors: The single continuous stroke is enforced by construction via optimization of a single URBS curve. However, we acknowledge that the current manuscript provides insufficient detail on initialization and regularization to prevent practical degeneracies. In the revision we will expand the method section to specify the initialization (coarse curve seeded from semantic guidance) and the regularization losses (penalizing excessive curvature and segment overlap) that maintain valid single-stroke outputs. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper introduces a novel optimization pipeline that applies the established score distillation sampling technique to the parameters of a uniform rational B-spline (URBS) curve representation, with added style losses, to generate single continuous-stroke vector drawings from text or image prompts. The continuity constraint is enforced directly by the chosen curve representation rather than derived from data or prior results. No load-bearing step reduces by construction to a fitted parameter, self-citation chain, or ansatz smuggled from the authors' prior work; the method is presented as an application of existing SDS to a new domain with downstream fabrication benefits. The central claims rest on empirical comparison to baselines and released code rather than any internal equivalence or uniqueness theorem imported from self-citation.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the effectiveness of score distillation sampling for guiding curve optimization and on the assumption that URBS can represent artist-style single-line drawings without additional invented entities.

axioms (2)
  • domain assumption Score distillation sampling can be applied to optimize geometric curve parameters to match semantic image distributions.
    Invoked in the abstract as the core optimization driver; no derivation supplied.
  • domain assumption A single URBS curve can capture the essential visual content of arbitrary subjects while remaining a continuous stroke.
    Stated as 'ensuring that the drawing consists of a single continuous stroke by design'.

pith-pipeline@v0.9.1-grok · 5715 in / 1251 out tokens · 21355 ms · 2026-06-28T11:52:52.737181+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

86 extracted references · 36 canonical work pages · 1 internal anchor

  1. [1]

    S., et al

    LeCun, Y. and Boser, B. and Denker, J. S. and Henderson, D. and Howard, R. E. and Hubbard, W. and Jackel, L. D. , title =. 1989 , issue_date =. doi:10.1162/neco.1989.1.4.541 , journal =

    work page doi:10.1162/neco.1989.1.4.541 1989

  2. [2]

    2025 , isbn =

    Arar, Ellie and Frenkel, Yarden and Cohen-Or, Daniel and Shamir, Ariel and Vinker, Yael , title =. 2025 , isbn =. doi:10.1145/3721238.3730612 , booktitle =

    work page doi:10.1145/3721238.3730612 2025

  3. [3]

    Image-driven Robot Drawing with Rapid Lognormal Movements , year =

    Berio, Daniel and Clivaz, Guillaume and Stroh, Michael and Deussen, Oliver and Plamondon, Réjean and Calinon, Sylvain and Leymarie, Frederic Fol , booktitle =. Image-driven Robot Drawing with Rapid Lognormal Movements , year =

  4. [4]

    2004 , issn =

    Continuous line drawings via the traveling salesman problem , journal =. 2004 , issn =. doi:https://doi.org/10.1016/j.orl.2003.10.001 , author =

    work page doi:10.1016/j.orl.2003.10.001 2004

  5. [5]

    Scaling rectified flow transformers for high-resolution image synthesis , year =

    Esser, Patrick and Kulal, Sumith and Blattmann, Andreas and Entezari, Rahim and M\". Scaling rectified flow transformers for high-resolution image synthesis , year =. Proceedings of the 41st International Conference on Machine Learning , articleno =

  6. [6]

    CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders , url =

    Frans, Kevin and Soros, Lisa and Witkowski, Olaf , booktitle =. CLIPDraw: Exploring Text-to-Drawing Synthesis through Language-Image Encoders , url =

  7. [7]

    Grattafiori, Aaron and Dubey, Abhimanyu and Jauhri, Abhinav and Pandey, Abhinav and Kadian, Abhishek and Al-Dahle, Ahmad and Letman, Aiesha and Mathur, Akhil and Schelten, Alan and Vaughan, Alex and Yang, Amy and Fan, Angela and Goyal, Anirudh and Hartshorn, Anthony and Yang, Aobo and Mitra, Archi and Sravankumar, Archie and Korenev, Artem and Hinsvark, A...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2407.21783

  8. [8]

    Edward J Hu and yelong shen and Phillip Wallis and Zeyuan Allen-Zhu and Yuanzhi Li and Shean Wang and Lu Wang and Weizhu Chen , booktitle =. Lo. 2022 , url =

    2022

  9. [9]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Jain, Ajay and Xie, Amber and Abbeel, Pieter , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2023 , pages =

    2023

  10. [10]

    and Bosch, Robert , title =

    Kaplan, Craig S. and Bosch, Robert , title =. Renaissance Banff: Mathematics, Music, Art, Culture , year =

  11. [11]

    2017 , eprint =

    Adam: A Method for Stochastic Optimization , author =. 2017 , eprint =

    2017

  12. [12]

    Optimal Transport Approximation of 2-Dimensional Measures , year =

    Lebrat, L\'. Optimal Transport Approximation of 2-Dimensional Measures , year =. doi:10.1137/18M1193736 , journal =

    work page doi:10.1137/18m1193736

  13. [13]

    ACM Trans

    Li, Tzu-Mao and Luk\'. Differentiable vector graphics rasterization for editing and learning , year =. doi:10.1145/3414685.3417871 , journal =

    work page doi:10.1145/3414685.3417871

  14. [14]

    Li, Hua and Mould, David , title =. Int. J. Creat. Interaces Comput. Graph. , month = jul, pages =. 2014 , issue_date =

    2014

  15. [15]

    2017 , isbn =

    Liu, Chenxi and Hodgins, Jessica and McCann, James , title =. 2017 , isbn =. doi:10.1145/3092919.3092925 , booktitle =

    work page doi:10.1145/3092919.3092925 2017

  16. [16]

    2002 , isbn =

    Secord, Adrian , title =. 2002 , isbn =. doi:10.1145/508530.508537 , booktitle =

    work page doi:10.1145/508530.508537 2002

  17. [17]

    InSpecial Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers ’24

    Tojo, Kenji and Shamir, Ariel and Bickel, Bernd and Umetani, Nobuyuki , title =. 2024 , isbn =. doi:10.1145/3641519.3657453 , booktitle =

    work page doi:10.1145/3641519.3657453 2024

  18. [18]

    InProceedings of the IEEE International Conference on Robotics and Automation (ICRA)

    Vinker, Yael and Pajouheshgar, Ehsan and Bo, Jessica Y. and Bachmann, Roman Christian and Bermano, Amit Haim and Cohen-Or, Daniel and Zamir, Amir and Shamir, Ariel , title =. 2022 , issue_date =. doi:10.1145/3528223.3530068 , journal =

    work page doi:10.1145/3528223.3530068 2022

  19. [19]

    Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =

    Vinker, Yael and Alaluf, Yuval and Cohen-Or, Daniel and Shamir, Ariel , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =. 2023 , pages =

    2023

  20. [20]

    Computer Graphics Forum , author =

    Wong, Fernando J. and Takahashi, Shigeo , title =. Computer Graphics Forum , volume =. doi:https://doi.org/10.1111/j.1467-8659.2011.02040.x , year =

    work page doi:10.1111/j.1467-8659.2011.02040.x 2011

  21. [21]

    and Takahashi, Shigeo , title =

    Wong, Fernando J. and Takahashi, Shigeo , title =. 2013 , issue_date =. doi:10.1007/s00371-013-0809-1 , journal =

    work page doi:10.1007/s00371-013-0809-1 2013

  22. [22]

    International Conference on Computer Graphics Theory and Applications , volume =

    Hierarchical design of continuous line illustrations , author =. International Conference on Computer Graphics Theory and Applications , volume =. 2013 , organization =

    2013

  23. [23]

    Proceedings of the 37th International Conference on Neural Information Processing Systems , articleno =

    Xing, Ximing and Wang, Chuang and Zhou, Haitao and Zhang, Jing and Yu, Qian and Xu, Dong , title =. Proceedings of the 37th International Conference on Neural Information Processing Systems , articleno =. 2023 , publisher =

    2023

  24. [24]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Xing, Ximing and Zhou, Haitao and Wang, Chuang and Zhang, Jing and Xu, Dong and Yu, Qian , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2024 , pages =

    2024

  25. [25]

    SVGDreamer++: Advancing Editability and Diversity in Text-Guided SVG Generation , year =

    Xing, Ximing and Yu, Qian and Wang, Chuang and Zhou, Haitao and Zhang, Jing and Xu, Dong , journal =. SVGDreamer++: Advancing Editability and Diversity in Text-Guided SVG Generation , year =

  26. [26]

    Depth Anything V2 , url =

    Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang , booktitle =. Depth Anything V2 , url =. doi:10.52202/079017-0688 , editor =

    work page doi:10.52202/079017-0688

  27. [27]

    Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =

    Zhang, Lvmin and Rao, Anyi and Agrawala, Maneesh , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =. 2023 , pages =

    2023

  28. [28]

    A Computational Approach to Edge Detection , year =

    Canny, John , journal =. A Computational Approach to Edge Detection , year =

  29. [29]

    2003 , issue_date =

    DeCarlo, Doug and Finkelstein, Adam and Rusinkiewicz, Szymon and Santella, Anthony , title =. 2003 , issue_date =. doi:10.1145/882262.882354 , journal =

    work page doi:10.1145/882262.882354 2003

  30. [30]

    Stress relief: improving structural strength of 3d printable objects,

    Eitz, Mathias and Hays, James and Alexa, Marc , title =. 2012 , issue_date =. doi:10.1145/2185520.2185540 , journal =

    work page doi:10.1145/2185520.2185540 2012

  31. [31]

    Proceedings of the 6th International Conference on Digital Arts (ARTECH) , pages =

    Hua Li and David Mould , title =. Proceedings of the 6th International Conference on Digital Arts (ARTECH) , pages =. 2012 , url =

    2012

  32. [32]

    Special Section on CANS: XDoG: An eXtended difference-of-Gaussians compendium including advanced image stylization , year =

    Winnem\". Special Section on CANS: XDoG: An eXtended difference-of-Gaussians compendium including advanced image stylization , year =. doi:10.1016/j.cag.2012.03.004 , journal =

    work page doi:10.1016/j.cag.2012.03.004 2012

  33. [33]

    2013 , issue_date =

    Akleman, Ergun and Xing, Qing and Garigipati, Pradeep and Taubin, Gabriel and Chen, Jianer and Hu, Shiyu , title =. 2013 , issue_date =. doi:10.1016/j.cag.2013.01.004 , journal =

    work page doi:10.1016/j.cag.2013.01.004 2013

  34. [34]

    and Maheswaranathan, Niru and Ganguli, Surya , title =

    Sohl-Dickstein, Jascha and Weiss, Eric A. and Maheswaranathan, Niru and Ganguli, Surya , title =. Proceedings of the 32nd International Conference on International Conference on Machine Learning - Volume 37 , pages =. 2015 , publisher =

    2015

  35. [35]

    Proceedings of the 31st International Conference on Neural Information Processing Systems , pages =

    Heusel, Martin and Ramsauer, Hubert and Unterthiner, Thomas and Nessler, Bernhard and Hochreiter, Sepp , title =. Proceedings of the 31st International Conference on Neural Information Processing Systems , pages =. 2017 , isbn =

    2017

  36. [36]

    International Conference on Learning Representations , year =

    A Neural Representation of Sketch Drawings , author =. International Conference on Learning Representations , year =

  37. [37]

    , title =

    Song, Jifei and Pang, Kaiyue and Song, Yi-Zhe and Xiang, Tao and Hospedales, Timothy M. , title =. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , month =

  38. [38]

    Photo-Sketching: Inferring Contour Drawings From Images , year =

    Li, Mengtian and Lin, Zhe and Mech, Radomir and Yumer, Ersin and Ramanan, Deva , booktitle =. Photo-Sketching: Inferring Contour Drawings From Images , year =

  39. [39]

    Proceedings of the 34th International Conference on Neural Information Processing Systems , articleno =

    Ho, Jonathan and Jain, Ajay and Abbeel, Pieter , title =. Proceedings of the 34th International Conference on Neural Information Processing Systems , articleno =. 2020 , isbn =

    2020

  40. [40]

    Deep Sketch-Based Modeling: Tips and Tricks , year =

    Zhong, Yue and Gryaditskaya, Yulia and Zhang, Honggang and Song, Yi-Zhe , booktitle =. Deep Sketch-Based Modeling: Tips and Tricks , year =

  41. [41]

    Srinivasan, Matthew Tancik, Jonathan T

    Mildenhall, Ben and Srinivasan, Pratul P. and Tancik, Matthew and Barron, Jonathan T. and Ramamoorthi, Ravi and Ng, Ren , title =. 2021 , issue_date =. doi:10.1145/3503250 , journal =

    work page doi:10.1145/3503250 2021

  42. [42]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Ribeiro, Leo Sampaio Ferraz and Bui, Tu and Collomosse, John and Ponti, Moacir , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

  43. [43]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Lin, Hangyu and Fu, Yanwei and Xue, Xiangyang and Jiang, Yu-Gang , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

  44. [44]

    Available: https://doi.org/10.1145/2897824.2925953

    Sangkloy, Patsorn and Burnell, Nathan and Ham, Cusuh and Hays, James , title =. 2016 , issue_date =. doi:10.1145/2897824.2925954 , journal =

    work page doi:10.1145/2897824.2925954 2016

  45. [45]

    Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) , month =

    Kampelmuhler, Moritz and Pinz, Axel , title =. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) , month =

  46. [46]

    2021 , issue_date =

    Yu, Chris and Schumacher, Henrik and Crane, Keenan , title =. 2021 , issue_date =. doi:10.1145/3439429 , journal =

    work page doi:10.1145/3439429 2021

  47. [47]

    Towards Practical Sketch-Based 3D Shape Generation: The Role of Professional Sketches , year =

    Zhong, Yue and Qi, Yonggang and Gryaditskaya, Yulia and Zhang, Honggang and Song, Yi-Zhe , journal =. Towards Practical Sketch-Based 3D Shape Generation: The Role of Professional Sketches , year =

  48. [48]

    2021 , isbn =

    V, Varshaneya and S, Balasubramanian and Balasubramanian, Vineeth , title =. 2021 , isbn =. doi:10.1145/3490035.3490258 , booktitle =

    work page doi:10.1145/3490035.3490258 2021

  49. [49]

    International conference on machine learning , pages =

    Learning transferable visual models from natural language supervision , author =. International conference on machine learning , pages =. 2021 , organization =

    2021

  50. [50]

    International Conference on Learning Representations , year =

    Score-Based Generative Modeling through Stochastic Differential Equations , author =. International Conference on Learning Representations , year =

  51. [51]

    The Eleventh International Conference on Learning Representations , year =

    DreamFusion: Text-to-3D using 2D Diffusion , author =. The Eleventh International Conference on Learning Representations , year =

  52. [52]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Chan, Caroline and Durand, Fr\'edo and Isola, Phillip , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2022 , pages =

    2022

  53. [53]

    2022 , eprint =

    Classifier-Free Diffusion Guidance , author =. 2022 , eprint =

    2022

  54. [54]

    The eleventh international conference on learning representations , year =

    Sketchknitter: Vectorized sketch generation with diffusion models , author =. The eleventh international conference on learning representations , year =

  55. [55]

    Transactions on Machine Learning Research , issn =

    Maxime Oquab and Timoth. Transactions on Machine Learning Research , issn =. 2024 , url =

    2024

  56. [56]

    Attention is All you Need , url =

    Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N and Kaiser, and Polosukhin, Illia , booktitle =. Attention is All you Need , url =

  57. [57]

    ACM Trans

    Iluz, Shir and Vinker, Yael and Hertz, Amir and Berio, Daniel and Cohen-Or, Daniel and Shamir, Ariel , title =. 2023 , issue_date =. doi:10.1145/3592123 , journal =

    work page doi:10.1145/3592123 2023

  58. [58]

    and Salimans, Tim and Poole, Ben and Ho, Jonathan , title =

    Kingma, Diederik P. and Salimans, Tim and Poole, Ben and Ho, Jonathan , title =. Proceedings of the 35th International Conference on Neural Information Processing Systems , articleno =. 2021 , isbn =

    2021

  59. [59]

    InSpecial Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers ’24

    Aigerman, Noam and Groueix, Thibault , title =. 2024 , isbn =. doi:10.1145/3641519.3657452 , booktitle =

    work page doi:10.1145/3641519.3657452 2024

  60. [60]

    2024 , isbn =

    Binninger, Alexandre and Sorkine-Hornung, Olga , title =. 2024 , isbn =. doi:10.1145/3680528.3687570 , booktitle =

    work page doi:10.1145/3680528.3687570 2024

  61. [61]

    Surface-Filling Curve Flows via Implicit Medial Axes , year =

    Noma, Yuta and Sell\'. Surface-Filling Curve Flows via Implicit Medial Axes , year =. doi:10.1145/3658158 , journal =

    work page doi:10.1145/3658158

  62. [62]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Youwang, Kim and Oh, Tae-Hyun and Pons-Moll, Gerard , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2024 , pages =

    2024

  63. [63]

    The Twelfth International Conference on Learning Representations , year =

    DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation , author =. The Twelfth International Conference on Learning Representations , year =

  64. [64]

    2025 , isbn =

    Zhan, Ruohao and Li, Yijin and He, Yisheng and Chen, Shuo and Shen, Yichen and Chen, Xinyu and Dong, Zilong and Huang, Zhaoyang and Zhang, Guofeng , title =. 2025 , isbn =. doi:10.1145/3746027.3755437 , booktitle =

    work page doi:10.1145/3746027.3755437 2025

  65. [65]

    Diffusion Model-Based Image Editing: A Survey , year =

    Huang, Yi and Huang, Jiancheng and Liu, Yifan and Yan, Mingfu and Lv, Jiaxi and Liu, Jianzhuang and Xiong, Wei and Zhang, He and Cao, Liangliang and Chen, Shifeng , journal =. Diffusion Model-Based Image Editing: A Survey , year =

  66. [66]

    Apparent ridges for line drawing , year =

    Judd, Tilke and Durand, Fr\'. Apparent ridges for line drawing , year =. doi:10.1145/1276377.1276401 , journal =

    work page doi:10.1145/1276377.1276401

  67. [67]

    Orientable Dense Cyclic Infill for Anisotropic Appearance Fabrication , year =

    Chermain, Xavier and Zanni, C\'. Orientable Dense Cyclic Infill for Anisotropic Appearance Fabrication , year =. doi:10.1145/3592412 , journal =

    work page doi:10.1145/3592412

  68. [68]

    2018 , issue_date =

    Zhao, Haisen and Zhang, Hao and Xin, Shiqing and Deng, Yuanmin and Tu, Changhe and Wang, Wenping and Cohen-Or, Daniel and Chen, Baoquan , title =. 2018 , issue_date =. doi:10.1145/3197517.3201338 , journal =

    work page doi:10.1145/3197517.3201338 2018

  69. [69]

    Worst-case structural analysis,

    Berger, Itamar and Shamir, Ariel and Mahler, Moshe and Carter, Elizabeth and Hodgins, Jessica , title =. 2013 , issue_date =. doi:10.1145/2461912.2461964 , journal =

    work page doi:10.1145/2461912.2461964 2013

  70. [70]

    OpenSketch: a richly-annotated dataset of product design sketches , year =

    Gryaditskaya, Yulia and Sypesteyn, Mark and Hoftijzer, Jan Willem and Pont, Sylvia and Durand, Fr\'. OpenSketch: a richly-annotated dataset of product design sketches , year =. doi:10.1145/3355089.3356533 , journal =

    work page doi:10.1145/3355089.3356533

  71. [71]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =

    Gal, Rinon and Vinker, Yael and Alaluf, Yuval and Bermano, Amit and Cohen-Or, Daniel and Shamir, Ariel and Chechik, Gal , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , month =. 2024 , pages =

    2024

  72. [72]

    ACM Trans

    Berio, Daniel and Stroh, Michael and Calinon, Sylvain and Fol Leymarie, Frederic and Deussen, Oliver and Shamir, Ariel , title =. 2025 , issue_date =. doi:10.1145/3763345 , journal =

    work page doi:10.1145/3763345 2025

  73. [73]

    2021 , issue_date =

    Mo, Haoran and Simo-Serra, Edgar and Gao, Chengying and Zou, Changqing and Wang, Ruomei , title =. 2021 , issue_date =. doi:10.1145/3450626.3459833 , journal =

    work page doi:10.1145/3450626.3459833 2021

  74. [74]

    2024 , issue_date =

    Choi, Changwoon and Lee, Jaeah and Park, Jaesik and Kim, Young Min , title =. 2024 , issue_date =. doi:10.1145/3658156 , journal =

    work page doi:10.1145/3658156 2024

  75. [75]

    Graph.37, 6, Article 203 (Dec

    Hsiao, Kai-Wen and Huang, Jia-Bin and Chu, Hung-Kuo , title =. 2018 , issue_date =. doi:10.1145/3272127.3275070 , journal =

    work page doi:10.1145/3272127.3275070 2018

  76. [76]

    Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =

    Liu, Richard and Fu, Daniel and Tan, Noah and Lang, Itai and Hanocka, Rana , title =. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , month =. 2025 , pages =

    2025

  77. [77]

    , title =

    Fu, Hongbo and Zhou, Shizhe and Liu, Ligang and Mitra, Niloy J. , title =. 2011 , issue_date =. doi:10.1145/2070781.2024167 , journal =

    work page doi:10.1145/2070781.2024167 2011

  78. [78]

    Curves and surfaces for CAGD: a practical guide , shorttitle =

    Farin, Gerald , year =. Curves and surfaces for CAGD: a practical guide , shorttitle =

  79. [79]

    Concorde

    Davauthor =. Concorde. 2015 , url =

    2015

  80. [80]

    2024 , url =

    StabilityAI , title =. 2024 , url =

    2024

Showing first 80 references.