| 410 | 0 | 37 |
| 下载次数 | 被引频次 | 阅读次数 |
人工智能的最新进展为风景园林学科带来了新的发展机遇。然而，在中国古典园林设计这类高语境领域中，通用模型仍面临数据集偏差与空间幻觉等问题。为克服单一平台方法在平衡文化语义与空间拓扑关系方面的局限，本文提出了一种逻辑驱动的多平台协同生成框架，将传统设计原则编码为可计算的约束条件。该框架将用于捕捉“意境”的语义发散机制与保障空间连贯性的拓扑控制方法相结合。研究采用弗雷歇初始距离、对比语言-预训练得分、峰值信噪比、结构相似性指数等指标，并结合专家打分法与分层消融实验对框架性能进行综合评价。结果表明，与单一平台的文生图基线方法（仅使用Midjourney）相比，完整的3层工作流使FID降低约30%，专家评定的文化真实性与空间逻辑性提升1.8～1.9分，同时将人工后期编辑时间减少约57%。分层消融实验显示，空间表现的提升主要归因于拓扑约束层，而文化真实性的增强则主要由本体优化层驱动。本研究既建立了一种可复现、与模型无关的文化遗产设计工作流，也在理论层面搭建了连接当前二维生成范式与未来三维空间认知之间的桥梁。
Abstract：Recent advances in artificial intelligence have introduced new opportunities for landscape architecture, yet high-context domains such as Chinese classical garden design remain challenged by dataset bias and spatial hallucination in general-purpose models. To overcome the limitations of single-platform approaches to balancing cultural semantics and spatial topology, we propose a logic-driven, multi-platform collaborative framework that encodes traditional design principles as computational constraints. The framework integrates semantic divergence to capture “yijing”(意境) with topological control to ensure spatial coherence. Performance is evaluated using metrics including Fréchet inception distance(FID), Contrastive Language-Image Pre-training, peak signal-tonoise ratio, and structural similarity index measure, together with expert assessments and ablation experiments. Results indicate that compared with a single-platform text-to-image baseline, the full three-layer workflow achieved about 30% reduction in FID and improved expert-rated cultural fidelity and spatial logic by approximately 1.8-1.9 points, while reducing human post-editing time by about 57%. Layer-wise ablation shows that spatial improvements are mainly attributable to the Topological Constraint Layer, whereas gains in cultural authenticity are driven by the Ontological Refinement Layer. This study not only establishes a reproducible, model-agnostic workflow for cultural heritage design but also provides a theoretical bridge connecting current 2D generative paradigms with future 3D spatial cognition.
[1]Tran,D.K.,Ignatieva,M.,&Tenorio,R.(2023).Exploring classical Vietnamese gardens under the Nguyen Dynasty:A comparative study with Chinese counterparts.Landscape Architecture Frontiers,11(5),44±58.
[2]Yang,H.(2022).A Treatise on the Garden of JiangnanÐA Study on the Art of Chinese Classical Garden.Springer.
[3]Guo,L.,Ma,W.,Gong,X.,Zhang,D.,Zhai,Z.,&Li,M.(2024).Digital preservation of classical gardens at the San Su Shrine.Heritage Science,12,66.
[4]Shan,P.,&Sun,W.(2021).Research on landscape design system based on 3D virtual reality and image processing technology.Ecological Informatics,63,101287.
[5]Mc Guire,J.,De Cremer,D.,&Van de Cruys,T.(2024).Establishing the importance of co-creation and self-efficacy in creative collaboration with artificial intelligence.Scientific Reports,14,18525.
[6]Huang,Y.-C.,Wensveen,S.,&Funk,M.(2023).Experiential speculation in vision-based AI design education:Designing conventional and progressive AI futures.International Journal of Design,17(2),1±17.
[7]Albukhari,I.N.(2025).The role of artificial intelligence(AI) in architectural design:A systematic review of emerging technologies and applications.Journal of Umm Al-Qura University for Engineering and Architecture,16,1457±1476.
[8]Tan,L.,&Luhrs,M.(2024).Using generative AI Midjourney to enhance divergent and convergent thinking in an architect’s creative design process.The Design Journal,27(4),677±699.
[9]Jiang,F.,Ma,J.,Webster,C.J.,Chiaradia,A.J.F.,Zhou,Y.,Zhao,Z.,&Zhang,X.(2024).Generative urban design:A systematic review on problem formulation,design generation,and decision-making.Progress in Planning,180,100795.
[10]Bao,Y.,&Xiang,C.(2024).Exploration of conceptual design generation based on the deep learning modelÐDiscussing the application of AIgenerator to the preliminary architectural design process.Creativity in the Age of Digital Reproduction,343,171±178.
[11]Paananen,V.,Oppenlaender,J.,&Visuri,A.(2023).Using text-to-image generation for architectural design ideation.International Journal of Architectural Computing,22(3),458±474.
[12]Chen,J.,Wang,D.,Shao,Z.,Zhang,X.,Ruan,M.,Li,H.,&Li,J.(2023).Using artificial intelligence to generate master-quality architectural designs from text descriptions.Buildings,13(9),2285.
[13]Alyıldız,E.C.(2023).Generative text-to-image models in architectural design:A study on relationship of language,architectural quality and creativity.Icontech International Journal of Surveys,Engineering,Technology,7(3),12±26.
[14]Al-Tabeeb,A.K.,&Al-Desouqi,A.(2023).Metaverse in architecture:An approach to documenting and exploring the Egyptian heritage through metaverse.Green Building&Construction Economics,4(2),276±295.
[15]Ahmed,H.(2023).The influence of Fatimid architecture heritage on sustainable design of contemporary architectural facades.International Design Journal,13(1),381±392.
[16]Chen,J.,Shao,Z.,&Hu,B.(2023).Generating interior design from text:Anew diffusion model-based method for efficient creative design.Buildings,13(7),1861.
[17]Chen,H.-C.,&Chen,Z.(2023).Using Chat GPT and Midjourney to generate Chinese landscape painting of Tang poem‘The Difficult Road to Shu’.International Journal of Social Sciences and Artistic Innovations,3(2),1±10.
[18]Jang,S.,Roh,H.,&Lee,G.(2025).Generative AI in architectural design:Application,data,and evaluation methods.Automation in Construction,174,106174.
[19]Li,C.,Zhang,T.,Du,X.,Zhang,Y.,&Xie,H.(2025).Generative AI models for different steps in architectural design:A literature review.Frontiers of Architectural Research,14(3),759±783.
[20]Zhou,S.,Jia,W.,Diao,H.,Geng,X.,Wu,Y.,Wang,M.,...Wu,Z.(2025).A Cycle GAN-Pix2pix framework for multi-objective 3D urban morphology optimization:Enhancing thermal performance in high-density areas.Sustainable Cities and Society,126,106400.
[21]Quan,S.J.(2022).Urban-GAN:An artificial intelligence-aided computation system for plural urban design.Environment and Planning B:Urban Analytics and City Science,49(9),2500±2515.
[22]Li,X.,Lu,W.,Peng,Z.,Zhang,Y.,&Huang,J.(2024).Generative design of walkable urban cool spots using a novel heuristic GAN x GAN approach.Building and Environment,266,112027.
[23]Zhuang,X.,Ju,Y.,Yang,A.,&Caldas,L.(2023).Synthesis and generation for 3D architecture volume with generative modeling.International Journal of Architectural Computing,21(2),297±314.
[24]Liu,J.,Qiu,Z.,Wang,L.,Liu,P.,Cheng,G.,&Chen,Y.(2024).Intelligent floor plan design of modular high-rise residential building based on graphconstrained generative adversarial networks.Automation in Construction,159,105264.
[25]Wang,Y.,Xie,L.,&Huang,M.(2026).Landscape design concept generation combining cultural mapping technology and multimodal modeling.Scientific Reports,16,1505.
[26]Zhou,H.,&Xiang,S.(2024).Applicability evaluation and reflection on artificial intelligence-basedªimage to imageºgeneration of landscape architecture masterplans.Landscape Architecture Frontiers,12(2),58±67.
[27]Ren,T.,Zeng,Z.,Hao,X.,&Zhang,X.(2025).A generative optimization framework for pocket parks supported by mature plant carbon sequestration data.Urban Forestry&Urban Greening,112,128993.
[28]Feng,L.,Sun,Y.,Yu,C.,Chen,R.,&Zhao,J.(2025).The applicability of two generative adversarial networks to generative plantscape design:Acomparative study.Land,14(4),746.
[29]Luo,Y.,He,J.,&He,Y.(2017).A rule-based city modeling method for supporting district protective planning.Sustainable Cities and Society,28,277±286.
[30]Wang,X.,Song,Y.,&Tang,P.(2020).Generative urban design using shape grammar and block morphological analysis.Frontiers of Architectural Research,9(4),914±924.
[31]Hanafy,N.O.(2023).Artificial intelligence’s effects on design process creativity:ªA study on used A.I.text-to-image in architectureº.Journal of Building Engineering,80,107999.
[32]Deng,Q.,Li,X.,Liu,Y.,&Hu,K.(2023).Exploration of three-dimensional spatial learning approach based on machine learning±Taking Taihu stone as an example.Architectural Intelligence,2,5.
[33]Aalaei,M.,Saadi,M.,Rahbar,M.,&Ekhlassi,A.(2023).Architectural layout generation using a graph-constrained conditional Generative Adversarial Network(GAN).Automation in Construction,155,105053.
[34]Li,Y.,Chen,H.,Yu,P.,&Yang,L.(2025).A review of artificial intelligence in enhancing architectural design efficiency.Applied Sciences,15(3),1476.
[35]Salma,Z.,Hijón-Neira,R.,&Pizarro,C.(2025).Designing co-creative systems:Five paradoxes in human±AI collaboration.Information,16(10),909.
[36]Yan,S.,Wu,C.,&Zhang,Y.(2025).Generative design for architectural spatial layouts:A review of technical approaches.Journal of Asian Architecture and Building Engineering,1±21.
[37]Ji,C.(2009).Yuan Ye[The craft of gardens].Chongqing Publishing House.
[38]Wen,Z.(2012).Zhang Wu Zhi[Treatise on superfluous things].Zhonghua Book Company.
[39]Chen,Y.,Gu,Y.,Liu,Y.,&Cao,L.(2024).Unveiling the dynamics ofªscenes changing as steps moveºin a Chinese classical garden:A case study of Jingxinzhai Garden.Heritage Science,12,131.
[40]Siu,V.M.(1999).China and Europe intertwined:A new view of the European sector of the Chang Chun Yuan.Studies in the History of Gardens&Designed Landscapes,19(3±4),376±393.
[41]Wang,Z.,Song,P.,Zhang,Q.,Wei,T.,&Pan,B.(2024).Digital improvements in the design and construction process of classical Chinese garden rockeries:A study based on material digitization.Heritage Science,12,327.
[42]Sun,M.,&Bao,Y.(2025).Effects of reflecting water on viewers’visual attention,spatial perception,and emotional responses:A case study in Chinese heritage garden.npj Heritage Science,13(1),26.
[43]Lin,S.,&Liu,W.(2024).Cultural connotation and bionic design application of foliage plants in Chinese gardens.Arts&Communication,2(1),1854.
[44]Zhu,Y.,Lee,S.-H.,&Choi,K.-R.(2022).Analysis of the space and design of Chinese,Japanese and Korean academy gardens.Land,11(9),1474.
(1) The spatial arrangement and combination of Chinese classical garden elements generate complex spatial and temporal relationships, including viewing and being viewed, primary and secondary elements with focal points, spatial contrasts,concealment and exposure, guidance and suggestion, sparsity and density,undulation and layering, void and solid, meandering paths, varied terrain, upward and downward perspectives, penetration and layering, and spatial sequences.
(2)“Yijing” refers to “意境“in Chinese, which might be translated as “imagery” in relevant literatures.
基本信息:
中图分类号:TP18;TU986
引用信息:
[1]许沉风,梁慧琳.AIGC赋能的中国古典园林的逻辑驱动生成框架：连接文化语义与空间约束（英文）[J].景观设计学(中英文),2026,14(04):89-103.
基金信息:
National Natural Science Foundation of China (No. 32371940); Humanities and Social Science Fund of Ministry of Education of China (No. 23YJCZH119)~~
2026-08-15
2026-08-15