[1]杨道俊,沈 雷*,杨芳宇.基于扩散模型的生成式人工智能服装设计[J].服装学报,2025,10(04):311-319.
　YANG Daojun,SHEN Lei*,YANG Fangyu.Diffusion Model-Based Generative Artificial Intelligence Generated Content for Fashion Design[J].Journal of Clothing Research,2025,10(04):311-319.
点击复制

基于扩散模型的生成式人工智能服装设计()
分享到：

《服装学报》[ISSN:2096-1928/CN:32-1864/TS]

卷:
第10卷
期数:
2025年04期
页码:
311-319
栏目:
智能服装
出版日期:
2025-09-13

文章信息/Info

Title:
Diffusion Model-Based Generative Artificial Intelligence Generated Content for Fashion Design
作者:
杨道俊1;  沈 雷*1;  杨芳宇2
1. 江南大学 设计学院,江苏 无锡 214122; 2. 厦门原壹创意设计有限公司,福建 厦门 361021
Author(s):
YANG Daojun1;  SHEN Lei*1;  YANG Fangyu2
1. School of Design, Jiangnan University, Wuxi 214122,China; 2. Xiamen Yuanyi Creative Design Co., Ltd., Xiamen 361021,China
分类号:
TS 941.26
文献标志码:
A
摘要:
人工智能领域的快速发展给服装设计带来了新的机遇与挑战。为了给服装设计从业者和研究人员提供新设计思路和研究方法,梳理了人工智能生成内容领域的最新研究成果,对其优化历程、应用领域进行剖析,并探讨神经对抗网络、Transformer、扩散模型、神经辐射场4类图像生成模型的优劣势。研究表明:扩散模型是服装设计领域优势较为突出的图像生成模型。基于稳态扩散模型并结合开源社区工具包进行大型预训练扩散模型的搭建,其模型及算法能够根据研究者的创作思路高效生成系列服装设计方案; 该模型生成的服装设计方案可通过模糊综合评价,验证其实际效果及潜在的应用价值。该研究能够为人工智能背景下服装设计体系提供理论基础,并进一步对未来人工智能生成内容在服装设计领域的应用提供实践基础。
Abstract:
The rapid development of artificial intelligence has brought new opportunities and challenges to the field of fashion design. To provide novel design ideas and research methods for fashion design practitioners and researchers, this study reviewed the latest research progress in the field of artificial intelligence-generated content(AIGC), analyzed its optimization process and application domains, and examined the strengths and weaknesses of four types of image generation models: Generative Adversarial Networks(GANs), Transformers, Diffusion Models, and Neural Radiance Fields(NeRF). The results demonstrate that diffusion models exhibit relatively outstanding advantages as image generation models in the context of fashion design. Based on the Stable Diffusion model and incorporating open-source community toolkits, a large-scale pre-trained diffusion model was constructed. The model and its associated algorithms can efficiently generate a series of fashion design schemes aligned with the creative intent of researchers. The design solutions produced by the model were subjected to a fuzzy comprehensive evaluation, which verified their practical effectiveness and potential application value. This research provides a theoretical foundation for fashion design systems in the context of artificial intelligence and further offers a practical basis for the future application of AIGC in the field of fashion design.

参考文献/References:

[1] 周愿恩. 图像描述生成中的若干关键技术研究[D]. 合肥: 合肥工业大学, 2022.
[2] 李白杨, 白云, 詹希旎, 等. 人工智能生成内容(AIGC)的技术特征与形态演进[J]. 图书情报知识, 2023, 40(1): 66-74.
LI Baiyang, BAI Yun, ZHAN Xini, et al. The technical features and aromorphosis of artificial intelligence generated content(AIGC)[J]. Documentation Information and Knowledge, 2023, 40(1): 66-74.(in Chinese)
[3] ZHAN F, YU Y, WU R, et al. Multimodal image synthesis and editing: a survey[J]. arXiv,2021:13592.
[4] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144.
[5] 施倩, 罗戎蕾. 基于生成对抗网络的服装图像生成研究进展[J]. 现代纺织技术, 2023, 31(2): 36-46.
SHI Qian, LUO Ronglei. Research progress of garment image generation based on generative adversarial network[J]. Modern Textile Technology, 2023, 31(2): 36- 46.(in Chinese)
[6] LIU K L, LI W, YANG C Y, et al. Intelligent design of multi-media content in Alibaba[J]. Frontiers of Information Technology and Electronic Engineering, 2019, 20(12): 1657-1664.
[7] 陈伟. 生成对抗网络在广告创意图制作中的应用[D]. 北京: 中国科学院大学, 2020.
[8] 杨怡然, 吴巧英. 智能化服装搭配推荐研究进展[J]. 浙江理工大学学报(自然科学版), 2021, 46(1): 1-12.
YANG Yiran, WU Qiaoying. Research progress of intelligent clothing matching recommendation[J]. Journal of Zhejiang Sci-Tech University(Natural Sciences Edition), 2021, 46(1): 1-12.(in Chinese)
[9] 杨争妍, 薛文良, 张传雄, 等. 基于生成式对抗网络的用户下装搭配推荐[J]. 纺织学报, 2021, 42(7): 164-168.
YANG Zhengyan, XUE Wenliang, ZHANG Chuanxiong, et al. Recommendations for user’s bottoms matching based on generative adversarial networks[J]. Journal of Textile Research, 2021, 42(7): 164-168.(in Chinese)
[10] 陈涵, 沈雷, 汪鸣明, 等. 基于生成对抗网络的书法纺织图案设计开发[J]. 丝绸, 2021, 58(2): 137-141.
CHEN Han, SHEN Lei, WANG Mingming, et al. Design and development of calligraphy textile patterns based on generative adversarial network[J]. Journal of Silk, 2021, 58(2): 137-141.(in Chinese)
[11] 曹宝秀. 用神经网络自动设计服装[J]. 中国纤检, 2019(4): 120-121.
CAO Baoxiu. An automatic garment designing by neural networks[J]. China Fiber Inspection, 2019(4): 120-121.(in Chinese)
[12] 张颖, 刘成霞. 生成对抗网络在虚拟试衣中的应用研究进展[J]. 丝绸, 2021, 58(12):63-72.
ZHANG Ying, LIU Chengxia. Research progress on the application of generative adversarial network in virtual fitting[J].Journal of Silk, 2011, 58(12): 63-72.(in Chinese)
[13] 徐小春. 深度学习算法在虚拟试衣中的应用[D]. 无锡: 江南大学, 2021.
[14] 王坤峰, 左旺孟, 谭营, 等. 生成式对抗网络: 从生成数据到创造智能[J]. 自动化学报, 2018, 44(5): 769-774.
WANG Kunfeng, ZUO Wangmeng, TAN Ying, et al. Generative adversarial networks: from generating data to creating intelligence[J]. Acta Automatica Sinica, 2018, 44(5): 769-774.(in Chinese)
[15] 雷新意, 吴陈. 一种基于BEGAN改进的残缺人脸图像修复算法[J]. 计算机与数字工程, 2023, 51(2): 484- 490.
LEI Xinxiang, WU Chen. A partial face image restoration algorithm based on BEGAN improvement [J]. Computer and Digital Engineering, 2023, 51(2): 484- 490.(in Chinese)
[16] BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[J]. Advances in neural information processing systems, 2020, 33: 1877-1901.
[17] CHEN M, RADFORD A, CHILD R, et al. Generative pretraining from pixels[C]//International conference on machine learning[S.I.]: PMLR, 2020: 1691-1703.
[18] WAN Z Y, ZHANG J B, CHEN D D, et al. High-fidelity plura-listic image completion with transformers[C]//2021 IEEE/CVF International Conference on Computer Vision.Montreal:IEEE, 2021: 4672- 4681.
[19] FENOCCHI E, MORELLI D, CORNIA M, et al. Dual-branch collaborative transformer for virtual try-on[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops(CVPRW). New Orleans: IEEE, 2022: 2246-2250.
[20] REN B, TANG H, MENG F Y, et al. Cloth interactive transformer for virtual try-on[J]. ACM Transactions on Multimedia Computing, Communications, and Applications, 2024, 20(4): 1-20.
[21] SHI J, ZHANG Y Y, WANG W H, et al. A novel two-stream transformer-based framework for multi-modality human action recognition[J]. Applied Sciences, 2023, 13(4): 2058.
[22] XIONG Z N, WANG C X, LI Y, et al. Swin-pose: swin transformer based human pose estimation[C]//2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval(MIPR). San Francisco: IEEE, 2022: 228-233.
[23] GOENKA S, ZHENG Z H, JAISWAL A, et al. FashionVLP: vision language transformer for fashion retrieval with feedback[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). New Orleans: IEEE, 2022: 14085-14095.
[24] BECATTINI F, TEOTINI F M, DEL BIMBO A. Transformer-based graph neural networks for outfit generation[J]. IEEE Transactions on Emerging Topics in Computing, 2024, 12(1): 213-223.
[25] JIANG J B, WANG T, YAN H, et al. ClothFormer: taming video virtual try-on in all module[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). New Orleans:IEEE, 2022: 10789-10798.
[26] CROITORU F A, HONDRU V, IONESCU R T, et al. Diffusion models in vision: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(9): 10850-10869.
[27] KAWAR B, ZADA S, LANG O, et al. Imagic: text-based real image editing with diffusion models[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Vancouver: IEEE, 2023: 6007-6017.
[28] XIE Y H, TAKIKAWA T, SAITO S, et al. Neural fields in visual computing and beyond[J]. Computer Graphics Forum, 2022, 41(2): 641- 676.
[29] MILDENHALL B, SRINIVASAN P P, TANCIK M, et al. NeRF: representing scenes as neural radiance fields for view synthesis[C]//Computer Vision-ECCV 2020. Cham: Springer, 2020: 405-421.
[30] ZHAN F, YU Y, WU R, et al. Multimodal image synthesis and editing: A survey[J]. arXiv,2021:13592.
[31] 邓钇敏, 张旭龙, 司世景, 等. 虚拟人形象合成技术综述[J]. 大数据, 2023, 9(3): 114-139.
DENG Yimin, ZHANG Xulong, SI Shijing, et al. Review on synthesis technology of Virtual Human image [J]. Big Data, 2023, 9(3): 114-139.(in Chinese)
[32] GUO Y D, CHEN K Y, LIANG S, et al. AD-NeRF: audio driven neural radiance fields for talking head synthesis[C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV). Montreal:IEEE, 2021: 5764-5774.
[33] GAO X J, YANG J L, KIM J, et al. MPS-NeRF: generalizable 3D human rendering from multiview images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025, 47(8): 6110-6121.
[34] ZHAO S, CHEN D, CHEN Y C, et al. Uni-controlnet: All-in-one control to text-to-image diffusion models[J]. arXiv,2023:16322.
[35] 郑凯,王菂. 人工智能在图像生成领域的应用——以Stable Diffusion和ERNIE-ViLG为例[J]. 科技视界, 2022(35): 50-54.
ZHENG Kai, WANG Di. Application of artificial intelligence in the field of image generation: a case study of Stable Diffusion and Erney-Vilg[J]. Science and Technology Vision, 2022(35): 50-54.(in Chinese)
[36] HU E J, SHEN Y, WALLIS P, et al. Lora: Low-rank adaptation of large language models[J]. arXiv,2021:09685.
[37] 朱伟明,于家蓓.基于微调扩散模型的智能化衬衫款式图生成[J].服装学报,2025,10(3):260-267.
ZHU Weiming,YU Jiabei.Intelligent shirt style drawing generation based on fine-tuned diffusion model[J].Journal of Clothing Research,2025,10(3):260-267.(in Chinese)
[38] ZHANG M, RAO A Y, AGRAWALA M. Adding conditional control to text-to-image diffusion models[J]. arXiv,2023:05543.
[39] CAO Z, HIDALGO G, SIMON T, et al. OpenPose: realtime multi-person 2D pose estimation using part affinity fields[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43(1): 172-186.
[40] MICHELUCCI U. An introduction to autoencoders[J]. arXiv.2022:03898.
(责任编辑:张 雪)

相似文献/References:

[1]朱伟明,于家蓓.基于微调扩散模型的智能化衬衫款式图生成[J].服装学报,2025,10(03):260.
　ZHU Weiming,YU Jiabei.Intelligent Shirt Style Drawing Generation Based on Fine-Tuned Diffusion Model[J].Journal of Clothing Research,2025,10(04):260.

更新日期/Last Update: 2025-08-30