The swift advancement of image generation technology has rendered the generation of batches of automatic fashion design images feasible. Nevertheless, the manner of extracting corresponding patterns promptly and precisely from the generated personalized fashion images persists as a challenge. This study has formulated a 2D–3D–2D pattern generation methodology. This approach initially employs a fine-tuned diffusion model to generate high-quality garment images, subsequently utilizes a large-scale image 3D reconstruction model to automatically effect 3D reconstruction of the garment images, and ultimately uses 3D model reverse engineering technology to convert the 3D garment model into the corresponding garment pattern. We generated a dress pattern as an exemplar to validate the method. The experimental results indicated that the method is capable of converting fashion images generated by the diffusion model into their corresponding 2D patterns. The average error between the 2D pattern and the 3D model was less than 5 mm, and the 3D virtual try-on results of the generated pattern conformed to the style of the generated image, thereby verifying the efficacy of the method. Compared with manual pattern drafting, the time consumption was significantly reduced, providing robust support for the automated coupling of fashion design and pattern production.
Sohl-DicksteinJWeissEAMaheswaranathanN, et al. Deep unsupervised learning using nonequilibrium thermodynamics. In: Proceedings of the 32nd international conference on machine learning (eds F Bach and D Blei), Lille, France, 6–11 July 2015, pp. 2256–2265. New York, USA: ACM.
2.
GoodfellowIPouget-AbadieJMirzaM, et al. Generative adversarial nets. In: Advances in neural information processing systems 27 (eds Z Ghahramani, M Welling, C Cortes, et al.), Montreal, Canada, 8–13 December 2013, pp. 2672–2680. Cambridge, USA: MIT Press.
3.
WuQZhuBYongB, et al.
ClothGAN: generation of fashionable Dunhuang clothes using generative adversarial networks. Connect Sci2021;
33: 341–358.
4.
RenYujiaChenLChenY.Garment style interactive design based on DCGAN algorithm. Wool Text J2022;
50: 93–97.
5.
TirtawanTSusantoEKZamanPL, et al.
Batik clothes auto-fashion using conditional generative adversarial network and U-Net. (EIConCIT)2021: 145–150.
6.
WangLLiuJChengG, et al. Constructing a personalized AI assistant for shear wall layout using Stable Diffusion. ArXiv, https://arxiv.org/abs/2305.10830 (2023, accessed 21 October 2024).
7.
WangCChungJ.Research on AI painting generation technology based on the [Stable Diffusion]. Int J Adv Smart Converg2023;
12: 90–95.
8.
CaoSChaiWHaoS, et al.
Difffashion: reference-based fashion design with structure-aware transfer by diffusion models. IEEE Trans Multimedia2024;
26: 3962–3975.
9.
YunchuYWeiyuanZ.Prototype garment pattern flattening based on individual 3D virtual dummy. Int J Cloth Sci Technol2007;
19: 334–348.
10.
AuCKMaYS.Garment pattern definition, development and application with associative feature approach. Comput Ind2010;
61: 524–531.
11.
DeitkeMSchwenkDSalvadorD, et al. Objaverse: a universe of annotated 3D objects. In: 2023 IEEE/CVF conference on computer vision and pattern recognition (eds xxxx), Vancouver, Canada, 17–24 June 2023, pp. 13142–13153. Piscataway, NJ: IEEE.
12.
DeitkeMLiuRWallingfordM, et al. Objaverse-xl: A universe of 10m+ 3d objects. Advances in Neural Information Processing Systems, 2023, 36: 35799–35813.
13.
WuJZhangCXueT, et al. Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling. In: NIPS’16: proceedings of the 30th international conference on neural information processing systems (eds DD Lee and U von Luxburg), Barcelona, Spain, 5–10 December 2016, pp. 82–90. Red Hook, USA: Curran Associates.
14.
WangZWangYChenY, et al. CRM: single image to 3D textured mesh with convolutional reconstruction model. In: European conference on computer vision (eds A Leonardis, E Ricci, S Roth, et al.), Nicosia, Cyprus, 2–3 April 2025, pp. 57–74. Switzerland: Springer, Cham.
15.
BossMHuangZVasishtaA, et al.
SF3D: stable fast 3D mesh reconstruction with UV-unwrapping and illumination disentanglement. ArXiv2024: arXiv:2408.00653.
16.
VuruskanABulgunE.Identification of female body shapes based on numerical evaluations. Int J Cloth Sci Technol2011;
23: 46–60.
17.
ChoiY LNamY J.Classification of upper lateral body shapes for the apparel industry. Hum Factor Ergon Man2010,
20: 378–390.
18.
SongYErmonS. Generative modeling by estimating gradients of the data distribution. ArXiv, https://arxiv.org/abs/1907.05600 (2019, accessed 18 September 2024).
19.
HoJJainAAbbeelP. Denoising diffusion probabilistic models. In: Advances in neural information processing systems 34 (eds M Ranzato, A Beygelzimer, Y Dauphin, et al.), pp. 6840–6851. Cambridge, USA: MIT Press, 2020.
20.
DhariwalPNicholA. Diffusion models beat GANs on image synthesis. In Advances in neural information processing systems 35 (eds M Ranzato, A Beygelzimer, Y Dauphin, et al.), 6–14 2021, pp. 8780–8794. Cambridge, USA: MIT Press.
21.
RombachRBlattmannALorenzD, et al. High-resolution image synthesis with latent diffusion models. In: 2022 IEEE/CVF conference on computer vision and pattern recognition (eds Lisa O’Conner), New Orleans, USA, 18–24 June 2022, pp. 10684–10695. Piscataway, NJ: IEEE.
22.
Jaruga-RozdolskaA.Artificial intelligence as part of future practices in the architect’s work: MidJourney generative tool as part of a process of creating an architectural form. Architectus2022;
3: 95–104.
23.
MirzaMOsinderoS. Conditional generative adversarial nets. In: Advances in neural information processing systems 27 (eds Z Ghahramani, M Welling, C Cortes, et al). Montreal, Canada, 8–13 2014, pp. 5767–5777. Cambridge, USA: MIT Press.
24.
CaoZSimonTWEIS E, et al. Realtime multi-person 2D pose estimation using part affinity fields. In: 2017 IEEE conference on computer vision and pattern recognition (eds Lisa O’Conner), Honolulu, USA, 21–26 July 2017, pp. 7291–7299. Piscataway, NJ: IEEE.
25.
CannyJ.A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell1986;
8: 679–698.
26.
HUEShenYWallisP, et al. LoRA: low-rank adaptation of large language models. ArXiv, https://arxiv.org/abs/2106.09685v2 (2021, accessed 20 October 2024).
27.
RuizNLiYJampaniV, et al. DreamBooth: fine tuning text-to-image diffusion models for subject-driven generation. In: 2023 IEEE/CVF conference on computer vision and pattern recognition (eds Lisa O’Conner), Vancouver, Canada, 17–24 June 2023, pp. 22500–22510. Piscataway, NJ: IEEE.
28.
GalRAlalufYAtzmonY, et al.
An image is worth one word: personalizing text-to-image generation using textual inversion. ArXiv2022: arXiv:2208.01618.
29.
BennisCVezienJMIglesiasG.Piecewise surface flattening for non-distorted texture mapping. ACM1991;
25: 237–246
30.
YuGXPatrikalakisNMMaekawaT.Optimal development of doubly curved surfaces. Comput Aid Geom Des2000;
17: 545–577.
31.
HindsBKMcCartneyJWoodsG.Pattern development for 3D surfaces. Comput Aid Des1991;
23: 583–592.
32.
KimSJeongYLeeY, et al.
3D pattern development of tight-fitting dress for an asymmetrical female manikin. Fibers Polym2010;
11: 142–146.
33.
YangYZhangW.Prototype garment pattern flattening based on individual 3D virtual dummy. Int J Clothing Sci Technol2007;
19: 334–348
34.
TaoXBruniauxP.Toward advanced three-dimensional modeling of garment prototype from draping technique. Int J Clothing Sci Technol2013;
25: 266–283.
35.
KulińskaMAbtewM ABruniauxP, et al.
Block pattern design system using 3D zoning method on digital environment for fitted garment. Text Res J2022;
92: 4978–4993.
36.
ShimadaTTadaY.Approximate transformation of an arbitrary curved surface into a plane using dynamic programming. Comput Aid Des1991;
23: 153–159.
37.
ChoYKomatsuTInuiS, et al.
Individual pattern making using computerized draping method for clothing. Text Res J2006;
76: 646–654.
38.
AonoMDentiPBreenDE, et al.
Fitting a woven cloth model to a curved surface: dart insertion. IEEE Comput Graph Appl1996;
16: 60–70.
39.
AonoMBreenDEWoznyMJ.Fitting a wovencloth model to a curved surface-mapping algorithms. Comput Aid Des1994;
26: 278–292.
40.
HongYZhangKGuJ, et al.
LRM: large reconstruction model for single image to 3D. ArXiv2023: arXiv:2311.04400.
41.
TochilkinDPankratzDLiuZ, et al.
TripoSR: fast 3D object reconstruction from a single image. ArXiv2024: arXiv:2403.02151.
42.
LiPLiuYLongX, et al.
Era3D: high-resolution multiview diffusion using efficient row-wise attention. ArXiv2024: arXiv:2405.11616.
43.
XuJChengWGaoY, et al.
InstantMesh: efficient 3D mesh generation from a single image with sparse-view large reconstruction models. ArXiv2024: arXiv:2404.07191.
44.
TangJChenZChenX, et al. LGM: Large multi-view Gaussian model for high-resolution 3D content creation. In: LeonardisARicciERothS, et al. (eds) Computer vision – ECCV 2024. ECCV 2024. Lecture notes in computer science, vol 15062.
Switzerland:
Springer, Cham, 2025, pp. 1–18.
45.
ZouZXYuZGuoYC, et al. Triplane meets Gaussian splatting: fast and generalizable single-view 3D reconstruction with transformers. In: 2024 IEEE/CVF conference on computer vision and pattern recognition (eds Cristina Ceballos), Seattle, USA, 16–22 June 2024, pp. 10324–10335. Piscataway, NJ: IEEE.
46.
LongXGuoYCLinC, et al. Wonder3D: single image to 3D using cross-domain diffusion. In: 2024 IEEE/CVF conference on computer vision and pattern recognition (eds Cristina Ceballos), Seattle, USA, 16–22 June 2024, pp. 9970–9980. Piscataway, NJ: IEEE.
47.
WuKLiuFCaiZ, et al. Unique3D: high-quality and efficient 3D mesh generation from a single image. ArXiv, https://arxiv.org/abs/2405.20343 (2024, accessed 23 August 2024).
48.
RadfordAKimJWHallacyC, et al. Learning transferable visual models from natural language supervision. Proceedings of the 38th international conference on machine learning. ICML, 2021: 8748–8763.
49.
PascualRMaizaASesma-SaraM, et al. Enhancing DreamBooth with LoRA for generating unlimited characters with Stable Diffusion. 2024 international joint conference on neural networks, Yokohama, Japan, 30 June–5 July 2024, pp. 1–8. Location: IEEE.
50.
GuYFangXDengX.A new Chinese landscape paintings generation model based on Stable Diffusion using DreamBooth. ArXiv2024: arXiv:2408.08561.
51.
JangWAgapitoL. CodeNeRF: disentangled neural radiance fields for object categories. In: 2021 IEEE/CVF international conference on computer vision (eds Lisa O’Conner), Montreal, Canada, 10–17 October 2021, pp. 12949–12958. Piscataway, NJ: IEEE.
52.
MüllerNSimonelliAPorziL, et al. AutoRF: learning 3D object radiance fields from single view observations. In: 2022 IEEE/CVF conference on computer vision and pattern recognition (eds Lisa O’Conner), New Orleans, USA, 18–24 June 2022, pp. 3971–3980. Piscataway, NJ: IEEE.
53.
CaronMTouvronHMisraI, et al. Emerging properties in self-supervised vision transformers. In: 2021 IEEE/CVF international conference on computer vision (eds Lisa O’Conner), Montreal, Canada, 10–17 October 2021, pp. 9650–9660. Piscataway, NJ: IEEE.
54.
LiJLiDXiongC, et al. BLIP: bootstrapping language-image pre-training for unified vision-language understanding and generation. ArXiv, https://arxiv.org/abs/2201.12086 (2022, accessed 16 September 2024).
55.
ParkJKoBJangH. StyleBoost: a study of personalizing text-to-image generation in any style using DreamBooth. 2023 14th international conference on information and communication technology convergence (ICTC). Red Hook, USA: Curran Associates, 2023, pp. 93–98.
56.
NathanaelO TLumentutJ SMuliawanN H, et al.
HYPNOS: highly precise foreground-focused diffusion finetuning for inanimate objects. ArXiv2024: arXiv:2410.14265.
