{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,21]],"date-time":"2026-03-21T07:47:48Z","timestamp":1774079268066,"version":"3.50.1"},"reference-count":40,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T00:00:00Z","timestamp":1773792000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Chinese font generation is important for digital typography, cultural preservation, and personalized user interfaces. However, existing methods often face challenges in maintaining structural consistency, supporting diverse stylistic variations, and achieving computational efficiency simultaneously, especially in cloud-based environments. A key application is bandwidth-efficient font delivery, where compact structural templates replace large font files for on-demand style customization. To address these issues, this paper proposes Stroke2Font\u2014a hierarchical vector model with AI-driven optimization for dynamic Chinese font generation. The core model decouples structural representation from style rendering through stroke element decomposition and B\u00e9zier curve parameterization. To further balance structural fidelity, style diversity, and real-time performance, we introduce a three-module optimization framework: (1) a reinforcement learning policy for dynamic selection of B\u00e9zier control parameters to minimize rendering latency; (2) a genetic algorithm for exploring style vector spaces and generating novel font variants; and (3) an adaptive complexity-aware optimization strategy that dynamically configures parameters based on character structural complexity. Experimental results on a dataset of 150 Chinese characters with 1123 stroke trajectories and 5287 feature points demonstrate that the adaptive complexity-aware optimization achieves the highest trajectory similarity of 65.2%, representing a 6.4% relative improvement over baseline (61.3%). The evaluation covers characters ranging from 1 to 18 strokes across 6 stroke types, with standard deviation reduced to \u00b15.7% (compared to \u00b16.5% baseline), indicating more consistent performance. Quantitative analysis confirms that the method generalizes effectively across varying character complexity, with the optimization showing stable improvement regardless of stroke count distribution. These results validate that Stroke2Font provides an effective solution for high-quality, efficient, and scalable Chinese font generation in cloud-based applications.<\/jats:p>","DOI":"10.3390\/a19030231","type":"journal-article","created":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T17:12:26Z","timestamp":1773853946000},"page":"231","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Stroke2Font: A Hierarchical Vector Model with AI-Driven Optimization for Chinese Font Generation"],"prefix":"10.3390","volume":"19","author":[{"given":"Qing-Sheng","family":"Li","sequence":"first","affiliation":[{"name":"School of Media and Information Engineering, Communication University of Zhejiang, Hangzhou 310018, China"},{"name":"Hangzhou Liangyun Intelligent Technology Co., Ltd., Hangzhou 310023, China"},{"name":"Tongxiang Institute, Communication University of Zhejiang, Tongxiang 314500, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yu-Lin","family":"Bian","sequence":"additional","affiliation":[{"name":"School of Communication and Journalism, Communication University of Zhejiang, Hangzhou 310018, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhen-Hui","family":"Chai","sequence":"additional","affiliation":[{"name":"School of Communication and Journalism, Communication University of Zhejiang, Hangzhou 310018, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,18]]},"reference":[{"key":"ref_1","first-page":"1","article-title":"An analysis of the historical origin, development and cultural inher-itance of Chinese characters","volume":"314","author":"Chen","year":"2022","journal-title":"Sinogram Cult."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Zhang, T., Liu, M., Yang, Y., Zuo, F., and Wang, G. (2025, January 19\u201321). Ancient handwritten Chinese character recogni-tion via multi-style attention and feature fusion. Proceedings of the 2025 5th International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC), Hefei, China.","DOI":"10.1109\/AIAHPC66801.2025.11290210"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"194","DOI":"10.1007\/s00530-024-01396-8","article-title":"Quality evaluation methods of handwritten Chinese characters: A comprehensive survey","volume":"30","author":"Chen","year":"2024","journal-title":"Multimed. Syst."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"3755","DOI":"10.1007\/s00371-023-03026-2","article-title":"Easy recognition of artistic Chinese calligraphic characters","volume":"39","author":"Yang","year":"2023","journal-title":"Vis. Comput."},{"key":"ref_5","unstructured":"(1980). Information Technology\u2014Chinese Ideogram Coded Character Set (Basic Set). Standard No. GB\/T 2312-1980. Available online: https:\/\/openstd.samr.gov.cn \/bzgk\/gb\/newGbInfo?hcno=5664A728BD9D523DE3B99BC37AC7A2CC."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"127105","DOI":"10.1016\/j.eswa.2025.127105","article-title":"Deep learning for Chinese font generation: A survey","volume":"276","author":"Wang","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"ref_7","first-page":"660","article-title":"Description and Generation of Chinese Outline Based on C-B\u00e9zier Curves","volume":"12","author":"Yang","year":"2000","journal-title":"J. Comput. Aided Des. Comput. Graph."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Lin, J.W., Hong, C.Y., Chang, R.I., Wang, Y.C., Lin, S.Y., and Ho, J.M. (2015, January 14\u201316). Complete font generation of Chinese char-acters in personal handwriting style. Proceedings of the 34th International Performance Computing and Communications Conference, Nanjing, China. Available online: https:\/\/ieeexplore.ieee.org\/document\/7410313.","DOI":"10.1109\/PCCC.2015.7410321"},{"key":"ref_9","first-page":"219","article-title":"A structure and style model for Chinese character dynamic generation","volume":"53","author":"Li","year":"2017","journal-title":"Acta Sci. Nat. Univ. Pekin."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"102876","DOI":"10.1016\/j.displa.2024.102876","article-title":"SPFont: Stroke potential features embedded GAN for Chinese calligraphy font generation","volume":"85","author":"Chen","year":"2024","journal-title":"Displays"},{"key":"ref_11","first-page":"145","article-title":"Self-supervised Chinese font generation based on square-block transformation","volume":"52","author":"Zeng","year":"2022","journal-title":"Sci. Sin."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"105305","DOI":"10.1016\/j.dsp.2025.105305","article-title":"FourCornerGAN: Glyph formation augmentation for unpaired Chinese font generation","volume":"165","author":"Wang","year":"2025","journal-title":"Digit. Signal Process."},{"key":"ref_13","first-page":"221","article-title":"Personalized Handwritten Chinese Character Generation Method for Unsupervised Image Translation","volume":"58","author":"Lu","year":"2022","journal-title":"Comput. Eng. Appl."},{"key":"ref_14","unstructured":"Chen, W., Zhu, G., Li, Y., Ji, Y., and Liu, C. (2025). Advancements in Chinese font generation since deep learning era: A survey. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1561\/2000000039","article-title":"Deep learning: Methods and applications","volume":"7","author":"Deng","year":"2014","journal-title":"Found. Trends\u00ae Signal Process."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"336","DOI":"10.1093\/llc\/fqab075","article-title":"A visualizing analysis of Chinese character processing in the past 40 years (1981\u20132020)","volume":"37","author":"Cheng","year":"2022","journal-title":"Digit. Scholarsh. Humanit."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Li, Y., and Li, Y. (2021). Design and implementation of handwritten Chinese character recognition method based on CNN and TensorFlow. Proceedings of the 2021 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), IEEE.","DOI":"10.1109\/ICAICA52286.2021.9498146"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"5305","DOI":"10.1007\/s11042-021-11777-0","article-title":"An unsupervised font style transfer model based on generative adversarial networks","volume":"81","author":"Zeng","year":"2022","journal-title":"Multimed. Tools Appl."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Chang, B., Zhang, Q., Pan, S., and Meng, L. (2018). Generating handwritten Chinese characters using CycleGAN. Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), IEEE.","DOI":"10.1109\/WACV.2018.00028"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Wen, Q., Li, S., Han, B., and Yuan, Y. (2021, January 20\u201324). ZiGAN: Fine-grained Chinese calligraphy font generation via a few-shot style transfer approach. Proceedings of the 29th ACM Inter-national Conference on Multimedia, Chengdu, China.","DOI":"10.1145\/3474085.3475225"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"954","DOI":"10.1016\/j.procs.2025.01.056","article-title":"Conditional GANs in Image-to-Image Translation: Improving Accuracy and Contextual Relevance in Diverse Datasets","volume":"252","author":"Gandhi","year":"2025","journal-title":"Procedia Comput. Sci."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"783","DOI":"10.1007\/s11042-017-5457-4","article-title":"Font generation based on least squares conditional generative adversarial nets","volume":"78","author":"Lin","year":"2019","journal-title":"Multimed. Tools Appl."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Zhu, J.Y., Park, T., Isola, P., and Efros, A.A. (2017, January 22\u201329). Unpaired image-to-image translation using cycle-consistent adversarial networks. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.244"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"6737","DOI":"10.1007\/s11042-020-09709-5","article-title":"An end-to-end model for Chinese calligraphy generation","volume":"80","author":"Zhou","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Liu, J., Zheng, S., and Cai, Q. (2024, January 20\u201322). Handwritten Chinese Character Recognition Based on Attention Mechanism. Proceedings of the 2024 3rd International Conference on Artificial Intelligence and Computer Information Technology (AICIT), Yichang, China.","DOI":"10.1109\/AICIT62434.2024.10730466"},{"key":"ref_26","first-page":"331","article-title":"The fourier descriptor based automatic generation method for multiple Chinese fonts","volume":"7","author":"Pan","year":"1996","journal-title":"J. Softw."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Isola, P., Zhu, J.Y., Zhou, T., and Efros, A.A. (2017, January 21\u201326). Image-to-image translation with conditional adversarial networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.632"},{"key":"ref_28","unstructured":"Tian, Y. (2026, March 01). zi2zi: Master Chinese Calligraphy with Conditional Adversarial Networks. Available online: https:\/\/kaonashi-tyc.github.io\/2017\/04\/06\/zi2zi.html."},{"key":"ref_29","unstructured":"Jiang, Y., Lian, Z., Tang, Y., and Xiao, J. (February, January 27). SCFont: Structure guided Chinese font generation via deep stacked networks. Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA."},{"key":"ref_30","unstructured":"Wu, S.J., Yang, C.Y., and Hsu, J.Y.J. (2020). CalliGAN: Style and structure aware Chinese calligraphy character generator. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Wang, Y.-H., Gao, Y., and Lian, Z.-H. (2020). Attribute2Font: Creating Fonts You Want from Attributes. arXiv.","DOI":"10.1145\/3386569.3392456"},{"key":"ref_32","unstructured":"Jiang, H., Yang, G., Huang, K., and Zhang, R. (2024). W-Net: One-shot arbitrary-style Chinese character generation with deep neural networks. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Ning, F. (2022). Multi-style Migration of Chinese Characters based on Self-attention Mechanism and StarGAN v2. Proceedings of the 2022 3rd International Conference on Computer Vision, Image and Deep Learning & International Conference on Computer Engineering and Applications (CVIDL & ICCEA), IEEE. Available online: https:\/\/ieeexplore.ieee.org\/document\/9825460?denied=.","DOI":"10.1109\/CVIDLICCEA56201.2022.9825460"},{"key":"ref_34","unstructured":"Carlier, A., Danelljan, M., Alahi, A., and Timofte, R. (2020). DeepSVG: A Hierarchical Generative Network for Vector Graphics Animation. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Reddy, P., Gharbi, M., Luk\u00e1\u010d, M., and Mitra, N.J. (2021). Im2Vec: Synthesizing Vector Graphics without Vector Supervision. arXiv.","DOI":"10.1109\/CVPR46437.2021.00726"},{"key":"ref_36","first-page":"153","article-title":"Study of feature weighted-based generation method for dian strokes of Chinese character","volume":"50","author":"Li","year":"2014","journal-title":"Acta Sci. Nat. Univ. Pekin."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Mousas, C., Seo, H., Thalmann, D., and Cordier, F. (2026). Intelligent Compilation System for Chinese Character Animation Based on Dynamic Data Sets. Computer Animation and Social Agents, Springer.","DOI":"10.1007\/978-981-95-0100-7"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Li, Z., Li, Q., and Guan, Y. (2019). A Chinese Character Generation Model for Cloud Infor-mation Security. Proceedings of the 2019 IEEE International Conferences on Ubiquitous Computing & Communications (IUCC) and Data Science and Computational Intelligence (DSCI) and Smart Computing, Networking and Services (SmartCNS), IEEE.","DOI":"10.1109\/IUCC\/DSCI\/SmartCNS.2019.00114"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Li, Q., Li, J.-P., and Chen, L. (2018, January 28\u201330). A Bezier Curve-Based Font Generation Algorithm for Character Fonts. Proceedings of the 2018 IEEE 20th International Conference on High Performance Computing and Communications; IEEE 16th International Conference on Smart City; IEEE 4th International Conference on Data Science and Systems (HPCC\/SmartCity\/DSS), Exeter, UK.","DOI":"10.1109\/HPCC\/SmartCity\/DSS.2018.00194"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1038\/nature14236","article-title":"Human-level control through deep reinforcement learning","volume":"518","author":"Mnih","year":"2015","journal-title":"Nature"}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/231\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,21]],"date-time":"2026-03-21T05:15:54Z","timestamp":1774070154000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/231"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,18]]},"references-count":40,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["a19030231"],"URL":"https:\/\/doi.org\/10.3390\/a19030231","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,18]]}}}