pith. sign in

arxiv: 2601.16672 · v2 · submitted 2026-01-23 · 💻 cs.CV

ReWeaver: Towards Simulation-Ready and Topology-Accurate Garment Reconstruction

Ming Li , Hui Shan , Kai Zheng , Chentao Shen , Siyu Liu , Yanwei Fu , Zhen Chen , Xiangru Huang This is my paper

Pith reviewed 2026-05-16 12:13 UTC · model grok-4.3

classification 💻 cs.CV
keywords garment reconstructionsewing patternstopology accuracymulti-view imagesphysical simulation3D garmentdigital avatars
0
0 comments X

The pith

ReWeaver reconstructs garments from four views by predicting seams, panels and their exact connections in both 2D patterns and 3D space.

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

ReWeaver takes as few as four RGB images of a clothed person and outputs a full 3D garment model together with its 2D sewing pattern, including every seam location and how fabric panels join. Unlike earlier methods that produce loose point clouds or splats, the output is a structured mesh whose seams and panels match the input photos exactly. This structure makes the garment ready for direct use in cloth simulators, virtual try-on systems, and robotic handling tasks. The system is trained on a new synthetic dataset of over 100,000 multi-view samples that include ground-truth 3D shapes and annotated sewing patterns.

Core claim

ReWeaver predicts seams and panels as well as their connectivities in both the 2D UV space and the 3D space from as few as four input views, yielding structured 2D--3D garment representations suitable for high-fidelity physical simulation.

What carries the argument

ReWeaver network that jointly regresses 2D sewing patterns and 3D garment geometry while enforcing topology consistency between the two domains.

If this is right

  • Reconstructed garments can be imported straight into physics engines without manual seam editing.
  • Digital clothing assets gain accurate drape and wrinkle behavior under simulation.
  • Robotic systems receive graspable, topology-correct garment models for manipulation planning.
  • Virtual try-on pipelines can use the 2D patterns to generate new fabric variants.

Where Pith is reading between the lines

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

  • Extending the same joint 2D-3D prediction to video sequences could support dynamic garment tracking.
  • The structured output format opens a path to automatic pattern editing for custom-fit clothing.
  • Similar topology-aware reconstruction might apply to other layered objects such as upholstered furniture once suitable training data exists.

Load-bearing premise

Synthetic training images with perfect seam labels are close enough to real photographs that the learned predictions transfer without systematic errors in seam placement or panel connectivity.

What would settle it

On a set of real multi-view photographs of actual garments, the predicted seam lines fail to align with the visible stitching within the image resolution.

Figures

Figures reproduced from arXiv: 2601.16672 by Chentao Shen, Hui Shan, Kai Zheng, Ming Li, Siyu Liu, Xiangru Huang, Yanwei Fu, Zhen Chen.

Figure 1
Figure 1. Figure 1: ReWeaver. From as few as four input views, ReWeaver reconstructs high-precision sewing patterns with complex topology together with their corresponding 3D geometry. The method outputs a unified 2D–3D garment representation, where each panel and edge is explicitly linked to its associated 3D points. This enables faithful, simulation-ready garment assets to be recovered from ordinary and sparse-view photogra… view at source ↗
Figure 2
Figure 2. Figure 2: Visualization of the terminologies used in this paper. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Pipeline of our method. Our VGGT-like image encoder extracts features from multi-view images (Section 3.2), which then interact with predefined patch and curve queries. In the 3D geometry and topology prediction module (Section 3.3), these queries and the image tokens pass through stacked self- and cross-attention blocks. The resulting tokens are decoded into 3D curves and patches. The same tokens are then… view at source ↗
Figure 4
Figure 4. Figure 4: Texture differences between GCD and GCD-TS. Us￾ing the same garment geometry, GCD textures reveal strong seam cues (e.g., highlighted regions), which are unrealistic and can lead to overfitting. GCD-TS replaces these with more realistic, diverse textures to improve generalization. 4.2. Experimental Setup Training Details. We randomly split the GCD-TS dataset into training, validation, and test sets with a … view at source ↗
Figure 5
Figure 5. Figure 5 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Topology and geometry refinement. Without topology refinement, redundant 3D curves cause incorrect 2D edges (black segment in the leftmost image). Topology refinement removes these redundancies, producing clean panel structures, while geom￾etry refinement enforces accurate, closed-loop boundaries suitable for triangulation [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison with AIpparel. We compare ReWeaver and AIpparel against ground truth. Each example shows the predicted and ground-truth 2D panels along with the resulting simulated meshes. ReWeaver yields more accurate panels and correspondingly improved simulation results [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Visualization of more results ditioned on primitive-type embeddings; 3) cross-attention to the image features; and 4) a final feed-forward network. 2D Pattern Prediction. We assign a weight of 300 to the edge geometry loss and 1×10−2 to the scale loss. Each edge is represented by 50 points that are uniformly sampled along its arc length. For every panel, we normalize its 2D coor￾dinates to the range [−1, 1… view at source ↗
read the original abstract

High-quality 3D garment reconstruction plays a crucial role in mitigating the sim-to-real gap in applications such as digital avatars, virtual try-on and robotic manipulation. However, existing garment reconstruction methods typically rely on unstructured representations, such as 3D Gaussian Splats, struggling to provide accurate reconstructions of garment topology and sewing structures. As a result, the reconstructed outputs are often unsuitable for high-fidelity physical simulation. We propose ReWeaver, a novel framework for topology-accurate 3D garment and sewing pattern reconstruction from sparse multi-view RGB images. Given as few as four input views, ReWeaver predicts seams and panels as well as their connectivities in both the 2D UV space and the 3D space. The predicted seams and panels align precisely with the multi-view images, yielding structured 2D--3D garment representations suitable for 3D perception, high-fidelity physical simulation, and robotic manipulation. To enable effective training, we construct a large-scale dataset GCD-TS, comprising multi-view RGB images, 3D garment geometries, textured human body meshes and annotated sewing patterns. The dataset contains over 100,000 synthetic samples covering a wide range of complex geometries and topologies. Extensive experiments show that ReWeaver consistently outperforms existing methods in terms of topology accuracy, geometry alignment and seam-panel consistency.

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

1 major / 1 minor

Summary. The paper proposes ReWeaver, a framework for topology-accurate 3D garment and sewing pattern reconstruction from as few as four multi-view RGB images. It predicts seams, panels, and connectivities in both 2D UV and 3D spaces to produce structured representations suitable for physical simulation. The authors introduce the synthetic GCD-TS dataset (>100k samples) for training and claim consistent outperformance over prior methods on topology accuracy, geometry alignment, and seam-panel consistency.

Significance. If the central claims hold with real-world validation, the work would be significant for closing the sim-to-real gap in digital avatars, virtual try-on, and robotic manipulation by delivering simulation-ready garment outputs. The GCD-TS dataset construction is a clear positive contribution that could support future research. However, the current evaluation scope limits immediate impact.

major comments (1)
  1. [Experimental evaluation] Experimental evaluation (abstract and §4): The central claim requires that ReWeaver produces topology-accurate, simulation-ready outputs from real RGB images. All described training and quantitative evaluation occur on the synthetic GCD-TS dataset; no real-image test set, domain-adaptation module, or sim-to-real metrics are presented. This directly undermines the assertion of suitability for practical inputs under real lighting, texture, and deformation variations.
minor comments (1)
  1. [Abstract] Abstract: The statement of 'consistent outperformance' would be strengthened by including one or two key quantitative metrics (e.g., topology error or seam consistency scores) with reference to the corresponding table or figure.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the major comment below and outline planned revisions to strengthen the manuscript.

read point-by-point responses

  1. Referee: Experimental evaluation (abstract and §4): The central claim requires that ReWeaver produces topology-accurate, simulation-ready outputs from real RGB images. All described training and quantitative evaluation occur on the synthetic GCD-TS dataset; no real-image test set, domain-adaptation module, or sim-to-real metrics are presented. This directly undermines the assertion of suitability for practical inputs under real lighting, texture, and deformation variations.

    Authors: We agree that real-world validation is important for supporting claims of practical applicability. Our quantitative results are performed exclusively on the synthetic GCD-TS dataset, which was rendered with photorealistic textures, lighting variations, and diverse garment topologies to approximate real capture conditions. The ReWeaver architecture itself operates on raw RGB inputs and does not embed synthetic-specific assumptions. In the revised manuscript we will add a dedicated subsection with qualitative results on real multi-view studio-captured garment images, include a limitations discussion on the sim-to-real gap, and moderate language in the abstract and introduction to accurately reflect the current evaluation scope. Quantitative topology metrics on real data remain infeasible without corresponding ground-truth sewing annotations. revision: partial

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper presents a learned neural framework (ReWeaver) trained end-to-end on an independently constructed synthetic dataset GCD-TS containing multi-view images, 3D geometries, and annotated sewing patterns. No equations, loss terms, or architectural choices in the abstract or described method reduce by construction to the target outputs (seam/panel predictions); the model learns mappings from data rather than defining them tautologically. No self-citations are invoked as load-bearing uniqueness theorems, and no ansatz or renaming of known results is presented as a derivation. The central claims rest on empirical training and held-out evaluation, which are statistically independent of the reported predictions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that a large synthetic dataset can train a model that generalizes to real images for topology and seam accuracy; no free parameters or invented entities are explicitly introduced in the abstract.

axioms (1)
  • domain assumption Synthetic multi-view images and annotated sewing patterns in GCD-TS are sufficiently realistic and diverse to enable generalization to real garments.
    Training and evaluation both rely on this transfer assumption.

pith-pipeline@v0.9.0 · 5559 in / 1320 out tokens · 35905 ms · 2026-05-16T12:13:50.390558+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

52 extracted references · 52 canonical work pages · 5 internal anchors

  1. [1]

    3D Jungle.https://3djungle.net. 6

    work page

  2. [2]

    GPT-4 Technical Report

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ah- mad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. Gpt-4 technical report.arXiv preprint arXiv:2303.08774,

    work page internal anchor Pith review Pith/arXiv arXiv

  3. [3]

    architextures.https://architextures.org. 6

    work page

  4. [4]

    Layer Normalization

    Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E Hin- ton. Layer normalization.arXiv preprint arXiv:1607.06450,

    work page internal anchor Pith review Pith/arXiv arXiv

  5. [5]

    Cloth3d: clothed 3d humans

    Hugo Bertiche, Meysam Madadi, and Sergio Escalera. Cloth3d: clothed 3d humans. InEuropean Conference on Computer Vision, pages 344–359. Springer, 2020. 2, 3

    work page 2020

  6. [6]

    Chat- garment: Garment estimation, generation and editing via large language models

    Siyuan Bian, Chenghao Xu, Yuliang Xiu, Artur Grigorev, Zhen Liu, Cewu Lu, Michael J Black, and Yao Feng. Chat- garment: Garment estimation, generation and editing via large language models. InProceedings of the Computer Vi- sion and Pattern Recognition Conference, pages 2924–2934,

    work page

  7. [7]

    Bedlam: A synthetic dataset of bodies exhibit- ing detailed lifelike animated motion

    Michael J Black, Priyanka Patel, Joachim Tesch, and Jin- long Yang. Bedlam: A synthetic dataset of bodies exhibit- ing detailed lifelike animated motion. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8726–8737, 2023. 6

    work page 2023

  8. [8]

    End-to- end object detection with transformers

    Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to- end object detection with transformers. InEuropean confer- ence on computer vision, pages 213–229. Springer, 2020. 4

    work page 2020

  9. [9]

    Pergamo: Personalized 3d garments from monocular video

    Andr ´es Casado-Elvira, Marc Comino Trinidad, and Dan Casas. Pergamo: Personalized 3d garments from monocular video. InComputer Graphics Forum, pages 293–304. Wiley Online Library, 2022. 2

    work page 2022

  10. [10]

    H Onan Demirel, Salman Ahmed, and Vincent G Duffy. Digital human modeling: a review and reappraisal of ori- gins, present, and expected future methods for representing humans computationally.International Journal of Human– Computer Interaction, 38(10):897–937, 2022. 2

    work page 2022

  11. [11]

    Complexgen: Cad reconstruction by b-rep chain complex generation.ACM Transactions on Graphics (TOG), 41(4):1–18, 2022

    Haoxiang Guo, Shilin Liu, Hao Pan, Yang Liu, Xin Tong, and Baining Guo. Complexgen: Cad reconstruction by b-rep chain complex generation.ACM Transactions on Graphics (TOG), 41(4):1–18, 2022. 5

    work page 2022

  12. [12]

    HyperNetworks

    David Ha, Andrew Dai, and Quoc V Le. Hypernetworks. arXiv preprint arXiv:1609.09106, 2016. 5

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  13. [13]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. InProceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 5

    work page 2016

  14. [14]

    Dresscode: Autoregressively sewing and gen- erating garments from text guidance.ACM Transactions on Graphics (TOG), 43(4):1–13, 2024

    Kai He, Kaixin Yao, Qixuan Zhang, Jingyi Yu, Lingjie Liu, and Lan Xu. Dresscode: Autoregressively sewing and gen- erating garments from text guidance.ACM Transactions on Graphics (TOG), 43(4):1–13, 2024. 2, 3, 4

    work page 2024

  15. [15]

    GPT-4o System Card

    Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perel- man, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Weli- hinda, Alan Hayes, Alec Radford, et al. Gpt-4o system card. arXiv preprint arXiv:2410.21276, 2024. 3

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  16. [16]

    Bcnet: Learning body and cloth shape from a single image

    Boyi Jiang, Juyong Zhang, Yang Hong, Jinhao Luo, Ligang Liu, and Hujun Bao. Bcnet: Learning body and cloth shape from a single image. InEuropean Conference on Computer Vision, pages 18–35. Springer, 2020. 2

    work page 2020

  17. [17]

    Generating datasets of 3d garments with sewing patterns.arXiv preprint arXiv:2109.05633, 2021

    Maria Korosteleva and Sung-Hee Lee. Generating datasets of 3d garments with sewing patterns.arXiv preprint arXiv:2109.05633, 2021. 3

    work page arXiv 2021

  18. [18]

    Neuraltailor: Recon- structing sewing pattern structures from 3d point clouds of garments.ACM Transactions on Graphics (TOG), 41(4):1– 16, 2022

    Maria Korosteleva and Sung-Hee Lee. Neuraltailor: Recon- structing sewing pattern structures from 3d point clouds of garments.ACM Transactions on Graphics (TOG), 41(4):1– 16, 2022. 2, 3

    work page 2022

  19. [19]

    Garment- code: Programming parametric sewing patterns.ACM Trans- actions on Graphics (TOG), 42(6):1–15, 2023

    Maria Korosteleva and Olga Sorkine-Hornung. Garment- code: Programming parametric sewing patterns.ACM Trans- actions on Graphics (TOG), 42(6):1–15, 2023. 2, 6

    work page 2023

  20. [20]

    Garmentcodedata: A dataset of 3d made-to-measure garments with sewing pat- terns

    Maria Korosteleva, Timur Levent Kesdogan, Fabian Kem- per, Stephan Wenninger, Jasmin Koller, Yuhan Zhang, Mario Botsch, and Olga Sorkine-Hornung. Garmentcodedata: A dataset of 3d made-to-measure garments with sewing pat- terns. InEuropean Conference on Computer Vision, pages 110–127. Springer, 2024. 3, 6

    work page 2024

  21. [21]

    The hungarian method for the assignment problem.Naval research logistics quarterly, 2(1-2):83–97,

    Harold W Kuhn. The hungarian method for the assignment problem.Naval research logistics quarterly, 2(1-2):83–97,

    work page

  22. [22]

    State of the art in telepresence

    Jason Lawrence, Ye Pan, Dan B Goldman, Rachel Mc- Donnell, Julie Robillard, Carol O’Sullivan, Yaser Sheikh, Michael Zollhoefer, and Jason Saragih. State of the art in telepresence. InACM SIGGRAPH 2022 Courses, pages 1–

    work page 2022

  23. [23]

    Codi- mensional incremental potential contact.ACM Transactions on Graphics, 40(4), 2021

    Minchen Li, Danny M Kaufman, and Chenfanfu Jiang. Codi- mensional incremental potential contact.ACM Transactions on Graphics, 40(4), 2021. 3

    work page 2021

  24. [24]

    Single view garment reconstruction using diffusion mapping via pattern coordinates

    Ren Li, Cong Cao, Corentin Dumery, Yingxuan You, Hao Li, and Pascal Fua. Single view garment reconstruction using diffusion mapping via pattern coordinates. InProceedings of the Special Interest Group on Computer Graphics and In- teractive Techniques Conference Conference Papers, pages 1–11, 2025. 2, 3

    work page 2025

  25. [25]

    Garma- genet: A multimodal generative framework for sewing pat- tern design and generic garment modeling.arXiv preprint arXiv:2504.01483, 2025

    Siran Li, Chen Liu, Ruiyang Liu, Zhendong Wang, Gaofeng He, Yong-Lu Li, Xiaogang Jin, and Huamin Wang. Garma- genet: A multimodal generative framework for sewing pat- tern design and generic garment modeling.arXiv preprint arXiv:2504.01483, 2025. 3, 7

    work page arXiv 2025

  26. [26]

    Dress-1- to-3: Single image to simulation-ready 3d outfit with dif- fusion prior and differentiable physics.arXiv preprint arXiv:2502.03449, 2025

    Xuan Li, Chang Yu, Wenxin Du, Ying Jiang, Tianyi Xie, Yunuo Chen, Yin Yang, and Chenfanfu Jiang. Dress-1- to-3: Single image to simulation-ready 3d outfit with dif- fusion prior and differentiable physics.arXiv preprint arXiv:2502.03449, 2025. 2, 3

    work page arXiv 2025

  27. [27]

    Diffavatar: Simulation-ready garment optimization with differentiable simulation

    Yifei Li, Hsiao-yu Chen, Egor Larionov, Nikolaos Sarafi- anos, Wojciech Matusik, and Tuur Stuyck. Diffavatar: Simulation-ready garment optimization with differentiable simulation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4368– 4378, 2024. 2, 3

    work page 2024

  28. [28]

    Ani- matable gaussians: Learning pose-dependent gaussian maps for high-fidelity human avatar modeling

    Zhe Li, Zerong Zheng, Lizhen Wang, and Yebin Liu. Ani- matable gaussians: Learning pose-dependent gaussian maps for high-fidelity human avatar modeling. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 19711–19722, 2024. 2

    work page 2024

  29. [29]

    Visual instruction tuning.Advances in neural information processing systems, 36:34892–34916, 2023

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning.Advances in neural information processing systems, 36:34892–34916, 2023. 3

    work page 2023

  30. [30]

    Towards garment sewing pattern reconstruc- tion from a single image.ACM Transactions on Graphics (TOG), 42(6):1–15, 2023

    Lijuan Liu, Xiangyu Xu, Zhijie Lin, Jiabin Liang, and Shuicheng Yan. Towards garment sewing pattern reconstruc- tion from a single image.ACM Transactions on Graphics (TOG), 42(6):1–15, 2023. 2, 3, 7

    work page 2023

  31. [31]

    Multimodal latent diffusion model for complex sewing pattern generation

    Shengqi Liu, Yuhao Cheng, Zhuo Chen, Xingyu Ren, Wen- han Zhu, Lincheng Li, Mengxiao Bi, Xiaokang Yang, and Yichao Yan. Multimodal latent diffusion model for complex sewing pattern generation. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 17640– 17650, 2025. 3

    work page 2025

  32. [32]

    Smpl: A skinned multi- person linear model

    Matthew Loper, Naureen Mahmood, Javier Romero, Gerard Pons-Moll, and Michael J Black. Smpl: A skinned multi- person linear model. InSeminal Graphics Papers: Pushing the Boundaries, Volume 2, pages 851–866. 2023. 6

    work page 2023

  33. [33]

    Xpbd: position-based simulation of compliant constrained dynamics

    Miles Macklin, Matthias M ¨uller, and Nuttapong Chentanez. Xpbd: position-based simulation of compliant constrained dynamics. InProceedings of the 9th International Confer- ence on Motion in Games, pages 49–54, 2016. 3

    work page 2016

  34. [34]

    Marvelous designer official website,

    Marvelous Designer. Marvelous designer official website,

    work page

  35. [35]

    Algorithms for the assignment and trans- portation problems.Journal of the society for industrial and applied mathematics, 5(1):32–38, 1957

    James Munkres. Algorithms for the assignment and trans- portation problems.Journal of the society for industrial and applied mathematics, 5(1):32–38, 1957. 6

    work page 1957

  36. [36]

    Aipparel: A multimodal foundation model for digital garments

    Kiyohiro Nakayama, Jan Ackermann, Timur Levent Kes- dogan, Yang Zheng, Maria Korosteleva, Olga Sorkine- Hornung, Leonidas J Guibas, Guandao Yang, and Gordon Wetzstein. Aipparel: A multimodal foundation model for digital garments. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 8138–8149, 2025. 2, 3, 7, 4

    work page 2025

  37. [37]

    DINOv2: Learning Robust Visual Features without Supervision

    Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023. 4, 6, 2

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  38. [38]

    Rec-mv: Reconstructing 3d dynamic cloth from monocular videos

    Lingteng Qiu, Guanying Chen, Jiapeng Zhou, Mutian Xu, Junle Wang, and Xiaoguang Han. Rec-mv: Reconstructing 3d dynamic cloth from monocular videos. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4637–4646, 2023. 2

    work page 2023

  39. [39]

    Gaussian garments: Reconstruct- ing simulation-ready clothing with photorealistic appearance from multi-view video.arXiv preprint arXiv:2409.08189,

    Boxiang Rong, Artur Grigorev, Wenbo Wang, Michael J Black, Bernhard Thomaszewski, Christina Tsalicoglou, and Otmar Hilliges. Gaussian garments: Reconstruct- ing simulation-ready clothing with photorealistic appearance from multi-view video.arXiv preprint arXiv:2409.08189,

    work page arXiv

  40. [40]

    Gaussian garments: Reconstruct- ing simulation-ready clothing with photorealistic appearance from multi-view video

    Boxiang Rong, Artur Grigorev, Wenbo Wang, Michael J Black, Bernhard Thomaszewski, Christina Tsalicoglou, and Otmar Hilliges. Gaussian garments: Reconstruct- ing simulation-ready clothing with photorealistic appearance from multi-view video. In2025 International Conference on 3D Vision (3DV), pages 1054–1063. IEEE, 2025. 3

    work page 2025

  41. [41]

    Garmentimage: Raster encoding of garment sewing patterns with diverse topologies

    Yuki Tatsukawa, Anran Qi, I-Chao Shen, and Takeo Igarashi. Garmentimage: Raster encoding of garment sewing patterns with diverse topologies. InProceedings of the Special Inter- est Group on Computer Graphics and Interactive Techniques Conference Conference Papers, pages 1–11, 2025. 2

    work page 2025

  42. [42]

    Vggt: Vi- sual geometry grounded transformer

    Jianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht, and David Novotny. Vggt: Vi- sual geometry grounded transformer. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 5294–5306, 2025. 2, 3, 4

    work page 2025

  43. [43]

    Detr3d: 3d object detection from multi-view images via 3d-to-2d queries

    Yue Wang, Vitor Campagnolo Guizilini, Tianyuan Zhang, Yilun Wang, Hang Zhao, and Justin Solomon. Detr3d: 3d object detection from multi-view images via 3d-to-2d queries. InConference on robot learning, pages 180–191. PMLR, 2022. 4

    work page 2022

  44. [44]

    Dressing avatars: Deep photorealistic appearance for physically sim- ulated clothing.ACM Transactions on Graphics (TOG), 41 (6):1–15, 2022

    Donglai Xiang, Timur Bagautdinov, Tuur Stuyck, Fabian Prada, Javier Romero, Weipeng Xu, Shunsuke Saito, Jing- fan Guo, Breannan Smith, Takaaki Shiratori, et al. Dressing avatars: Deep photorealistic appearance for physically sim- ulated clothing.ACM Transactions on Graphics (TOG), 41 (6):1–15, 2022. 2

    work page 2022

  45. [45]

    Econ: Explicit clothed humans optimized via normal integration

    Yuliang Xiu, Jinlong Yang, Xu Cao, Dimitrios Tzionas, and Michael J Black. Econ: Explicit clothed humans optimized via normal integration. InProceedings of the IEEE/CVF con- ference on computer vision and pattern recognition, pages 512–523, 2023. 2

    work page 2023

  46. [46]

    Meta-learning via hypernetworks

    Dominic Zhao, Seijin Kobayashi, Jo ˜ao Sacramento, and Jo- hannes von Oswald. Meta-learning via hypernetworks. In 4th Workshop on Meta-Learning at NeurIPS 2020 (Met- aLearn 2020). NeurIPS, 2020. 5

    work page 2020

  47. [47]

    Physa- vatar: Learning the physics of dressed 3d avatars from visual observations

    Yang Zheng, Qingqing Zhao, Guandao Yang, Wang Yi- fan, Donglai Xiang, Florian Dubost, Dmitry Lagun, Thabo Beeler, Federico Tombari, Leonidas Guibas, et al. Physa- vatar: Learning the physics of dressed 3d avatars from visual observations. InEuropean Conference on Computer Vision, pages 262–284. Springer, 2024. 2

    work page 2024

  48. [48]

    sub-curve

    Xingxing Zou, Xintong Han, and Waikeung Wong. Cloth4d: A dataset for clothed human reconstruction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 12847–12857, 2023. 2 ReWeaver: Towards Simulation-Ready and Topology-Accurate Garment Reconstruction Supplementary Material A. Topology Refinement To obtain clean se...

    work page 2023

  49. [49]

    threshold-based filtering only

    work page

  50. [50]

    thresholding + 2D loop pruning

    work page

  51. [51]

    thresholding + loop pruning + sub-curve removal

    work page

  52. [52]

    The results in Table 3 show that each refinement stage con- tributes positively to edge accuracy and panel IoU

    thresholding + loop pruning + sub-curve removal + du- plicate merging (full refinement). The results in Table 3 show that each refinement stage con- tributes positively to edge accuracy and panel IoU. Because topology refinement has negligible influence on the Cham- fer Distance of patches and curves, we omit those metrics here. B. 2D Geometry Refinement ...

    work page