在近年來興起的生成式人工智能(AI)熱潮中,迅速發展、升級的圖像生成技術和制圖工具不斷沖擊著傳統景觀設計行業的工作模式。
導讀
人工智能(AI)圖像生成技術正在改變景觀設計中的傳統工作模式,其中,“圖生圖”式生成對抗網絡(GAN)技術具備輔助方案設計的潛能,因此面向用戶端對其展開技術適用性評價研究對于優化工具選擇、提升設計效率尤為重要。本研究旨在借助圖像分析和用戶調查方法,評估GAN生成方法生成結果的質量、與設計工作對接的有效性,以及景觀設計師對圖像生成結果的接受度。研究以Pix2Pix–BicycleGAN工作流中布局生成與平面渲染兩項任務為評價對象,建立了基于地塊數量的絕對/歐式距離、直方圖距離、結構相似性指數等圖像分析指標;針對GAN生成結果的視覺真實性和色彩肌理偏好開展了兩項在線用戶問卷調查。結果顯示,GAN生成布局與真實布局相似性高,GAN渲染平面能夠滿足概念方案呈現要求、用戶接受度好。最后,本文探討了GAN生成方法的內在合理性及其在行業倫理及數據偏見方面的局限性,反思現階段連接AI輔助設計與循證設計之間的技術空缺。
關鍵詞
景觀設計學;圖像生成;生成對抗網絡;人工智能輔助設計;適用性評價;景觀平面
人工智能“圖生圖”式
景觀平面生成技術的適用性評價與反思
Applicability Evaluation and Reflection on Artificial Intelligence-based “Image to Image Generation of Landscape Architecture Masterplans
在近年來興起的生成式人工智能(AI)熱潮中,迅速發展、升級的圖像生成技術和制圖工具不斷沖擊著傳統景觀設計行業的工作模式。目前可以對接景觀設計工作流的圖像生成技術主要被應用于平面生成及效果圖渲染兩方面。
平面生成的相關研究主要基于“圖生圖”式生成對抗網絡(GAN)開展。這類工具以建筑戶型平面生成為起點,目前已經發展至建筑排列方式與體塊關系的生成。近年來,景觀設計領域也開啟了平面生成的研究,但仍存在以下問題——缺乏公開可獲取的景觀平面數據集,訓練數據豐富性較低;可生成平面的尺度有限,主要適用于中小型綠地;針對GAN所生成平面的系統化定量評價較少,缺乏便于操作的評價指標;針對用戶端開展的調查較少,難以獲取使用評價。
效果圖生成的相關研究與應用主要圍繞Midjourney和Stable Diffusion兩大“文字生圖”(text to image)工具開展。相比之下,開源的Stable Diffusion模型除了可以通過關鍵詞生成圖片外,還具備“圖生圖”(image to image)和“模型生圖”(model to image)的自由訓練功能,目前,基于Stable Diffusion的建筑形體構思和建模工作流已經初步形成。
本研究關注基于GAN的景觀平面生成方法,從景觀設計師的視角綜合評估其技術適用性,以期為設計師在選擇工具時提供決策依據;旨在借助圖像分析和用戶調查方法,評估GAN生成方法生成結果的質量、與設計工作對接的有效性,以及景觀設計師對圖像生成結果的接受度。
本研究著眼于Pix2Pix–BicycleGAN景觀平面生成工作流中兩項關鍵任務——布局生成與平面渲染——的適應性評價。GAN生成的布局類似設計教學中的功能泡泡和平面草圖,是設計迭代和調整的基礎;GAN渲染圖則為布局中抽象的形態添加了色彩和肌理細節而使其更具可讀性。任務實現工具Pix2Pix是GAN領域應用較為廣泛的模型,而BicycleGAN是CycleGAN的改進模型。由于數據集中獲取與標注的平面類型有限,這一工作流目前主要適用于中小尺度的景觀場地。
Pix2Pix–BicycleGAN工作流中的布局生成與平面渲染示例 ? 周懷宇,向雙斌
通過向Pix2Pix模型輸入場地范圍,可以生成多種風格且包含不同用地類型的場地布局。評價圍繞生成的用地地塊布局與真實布局的相似性和視覺真實性開展。
在本研究中共收集2725張真實景觀平面圖,其中混合、曲線、折線、有機混合訓練集分別為2670、916、770、954張,預留用于評估生成效果的驗證集85張。基于4種樣式風格,共得到340個GAN生成布局用于后續評價。設計師在比較多個GAN生成布局后,依據項目需求并結合個人經驗形成更為精準的地塊布局,并將其作為平面渲染任務的輸入。
設計師將調整后的布局輸入BicycleGAN中,可獲取不同色彩肌理的渲染平面,方便與業主快速溝通設計思路。該任務的評價主要圍繞GAN渲染平面與人工渲染平面的相似性及用戶色彩肌理偏好開展。數據集共包含景觀平面325張,其中訓練集300張,驗證集25張。每張布局挑選暖色調、冷色結果各一張,評價總量為50張。
所生成五類用地的地塊數量(BN)能夠最直觀地反映GAN生成布局的形態多樣性,相應的地塊數量距離(BND)可用于評估340張由Pix2Pix生成的驗證集布局和真實布局的差異。其中,BND評價包含絕對BND和歐氏BND兩項指標的計算。本研究通過絕對距離比較單一樣式風格下生成布局中各類用地BN與真實布局之間的差值。同時,本研究通過絕對BND與歐式BND的聚合分析比較了四類樣式風格之間地塊劃分聚集程度的差異,并以聚合圖呈現兩組數據的中點聚集區。
圖像直方圖可顯示圖像中不同RGB像素的頻率分布,直方圖距離(HistD)則是衡量兩幅圖像之間像素分布差異的關鍵指標,能夠有效評估GAN生成布局與真實布局在用地地塊劃分與面積比例上的差異。其中,HistD的取值范圍為[0,1],取值小于0.5代表二者總體呈現相似趨勢。
地塊數量距離與直方圖距離方法示意 ? 周懷宇,向雙斌
結構相似性指數(SSIM)是一種廣泛使用的圖像相似性度量工具,可以評估兩幅經過不同處理加工的同源圖像(x, y)之間的感知差異。本研究通過計算SSIM來評估渲染平面與景觀設計師人工渲染平面的差別。SSIM的取值范圍為[0,1],其中,1表示兩幅圖像具有相同的結構,0則表示完全不同。此外,上述HistD指標也被并納入平面渲染評價指標。
為了評價GAN生成布局在視覺上能否以假亂真,同時了解從業人員的色彩肌理偏好,研究團隊于2023年9月1日至10月31日,面向景觀設計及相關領域的教師、學生和職業設計師發布了兩項問卷星在線調查問卷。問卷主要被投放到湖南大學建筑與規劃學院、清華大學建筑學院及北京市市政工程設計研究總院,同時要求受訪者選擇其求學或從業年限以確保結果的代表性與可靠性。
問卷1旨在對GAN生成布局進行圖靈測試,并評估從業人員對GAN生成布局的接受度。問卷中共涉及16張隨機抽取自驗證集的Pix2Pix自動生成布局,14張由知名事務所或大師創作方案的布局改繪,受訪者需要從中選出他們所認為的由AI生成的圖片,問卷并未設置最多可選數量限制。
在線調查問卷1(橙色編號代表GAN生成布局) ? 周懷宇,向雙斌
問卷2旨在判斷從業人員對幾類主流GAN模型渲染圖接受度的差異。問卷提供了30張渲染平面圖(10組、每組3張,分別來自Pix2pix、CycleGAN和BicycleGAN),要求受訪者判斷渲染圖是否達到在概念設計階段用于方案交流的標準,并依據色彩和肌理選擇每組中效果最佳的平面。
總體對比GAN生成布局與真實布局發現,兩者在圖形統計意義上的BN多樣性水平接近,地塊面積比例相似性突出。
1)由絕對BND平均值計算結果(表1)可知,單張布局中,GAN生成的五類用地BN與真實布局的差別均小于5個,主要差別體現在小品構筑物的數量上,這表明GAN與設計師在用地劃分時表現出的多樣性較為相似。
2)為了確定4類樣式訓練集的布局數量差異是否會導致BND結果的顯著不同,進一步對四類風格、五類用地地塊的絕對BND與歐氏BND進行聚合分析。結果顯示,4種樣式風格在用地劃分上也具有較強的相似性,進而可知訓練集數量的不同并沒有顯著影響訓練結果。
GAN生成布局與真實布局的絕對BND與歐氏BND聚合圖 ? 周懷宇,向雙斌
3)混合、曲線、直線、有機四類樣式風格的平均HistD值分別為0.41、0.45、0.41、0.43,均小于0.5,意味著GAN生成布局對不同用地類型劃分的面積比例總體與真實布局呈現接近趨勢。
計算50張渲染圖的平均SSIM和HistD值,結果如表2所示。總體來說,分析結果表明GAN渲染平面在像素分布、結構、對比度和亮度方面與職業設計師繪制的渲染圖高度相似。
問卷1共收到192份有效回復,55%的受訪者有5年以上的從業經歷,保證了結果的可靠性。結果顯示,16張GAN生成布局被識別為AI生成的平均概率為54.7%,略高于隨機猜測的概率。而GAN生成布局有約45%的幾率被從業人員錯認為是設計師創作的布局。同時,設計師創作的真實布局有約25%的概率被判定為GAN生成。總體而言,GAN生成布局可以使一些受訪者感到迷惑,同時,約70%的受訪者認為GAN技術有助力方案設計的潛力。
GAN生成布局與真實布局被識別為AI生成的平均概率 ? 周懷宇,向雙斌
研究進一步通過電話、微信、郵件等形式與受訪者交流如何辨別布局是由AI生成還是設計師繪制,發現功能設計中不合理的細節會嚴重破壞GAN方案的視覺真實性。研究將GAN布局缺陷分為三類:1)入口不完整;2)道路不連貫;3)節點不可達。其中,道路不連貫的問題最為明顯。
GAN生成布局的三類缺陷示例及設計師調整方案 ? 周懷宇,向雙斌
問卷2共收到422份有效問卷,受訪者中55%具有景觀設計專業背景, 37%位具有5年以上的從業經歷。結果顯示,91%的受訪者認為GAN平面渲染的質量可以滿足概念設計階段的方案推敲與溝通,47%的受訪者認為BicycleGAN在色彩和肌理效果上表現最佳。
本文通過引入圖像分析及用戶調查指標來評估GAN生成方法的技術適用性,旨在填補現有研究主要關注訓練方法而缺少后期評估的空白,為“圖生圖”生成式設計研究提供易于操作的評價框架。圖像分析結果顯示,GAN生成布局與真實布局的用地分布多樣性、渲染平面圖與設計師渲染平面圖的相似性均達到了較高的水平;用戶調查結果顯示,GAN生成布局具有較強的迷惑性、真假難辨,且渲染色彩和肌理得到了景觀設計師的認可。即使GAN生成方法模型內部存在較多的黑箱過程,但本研究為其內在邏輯的合理性提供了定量化支撐。
本研究的局限性主要體現在以下幾個方面。首先,研究未涉及對GAN生成方法的倫理評價。通常而言,設計需要基于特定的地域環境背景來實現功能需求,而GAN生成方法往往對復雜歷史、文化因素影響下的形式符號缺乏理解。本研究的問卷調查缺少對GAN生成方法原創性的關注,需在未來的研究中補充收集用戶對倫理問題的看法。其次,在本研究建立的評價框架中,未納入對GAN生成布局多樣性和訓練數據偏見問題的考量。AI工具輸出的內容受其訓練數據影響顯著,而目前景觀平面數據集多樣性嚴重不足,盲目應用會導致設計成果的同質化,未來亟需探索如何避免潛在的設計多樣性缺失。再者,本研究評價的Pix2Pix-BicycleGAN工作流雖然具有一定的典型性,但尚不能代表最前沿的技術迭代。在未來的研究中,可探索針對特定區域或類型景觀設計(如中國古典園林與西方現代景觀)的定制化GAN模型,在模型訓練過程中融入具有更多地域特征的數據,以及開發能夠辨識并強調這些特征的算法。
此外,GAN生成方法較低的可解釋性使其面臨著來自循證設計的挑戰。形態只是設計的一方面,而“設計結合自然”的科學思維要求綜合疊加各項因子(如豎向、土壤、徑流和植被等)以論證設計決策的合理性。因而,如何連接GAN模型代表的形態表達與物理模型代表的定量分析是AI深度融入設計學科必然要克服的問題。隨著GAN生成布局多樣性的提升,未來利用多目標優化算法對其進行篩選、優化將有助于提升設計決策的科學性。而隨著生成算法的更新,物理模型及優化算法將有可能逐步與AI模型融合,顯著提升GAN生成方法的可解釋性和應用深度。
[1] Bao, R. (2019). Research on intellectual analysis and application of landscape architecture based on machine learning. Landscape Architecture, 26(5), 29–34.
[2] Zhao, J., & Cao, Y. (2020). Review of artificial intelligence methods in landscape architecture. Chinese Landscape Architecture, 36(5), 82–87.
[3] Zhao, J., Chen, R., Hao, H., & Shao, Z. (2021). Application progress and prospect of machine learning technology in landscape architecture. Journal of Beijing Forestry University, 43(11), 137–156.
[4] Huang, W., & Zheng, H. (2018). Architectural Drawings Recognition and Generation Through Machine Learning. In: P. Anzalone, M. D. Signore, & A. J. Wit (Eds.), Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (pp. 18–20). ARCADIA.
[5] Nauata, N., Chang, K. H., Cheng, C. Y., Mori, G., & Furukawa, Y. (2020). House-GAN: Relational generative adversarial networks for graph-constrained house layout generation. In: A. Vedaldi, H. Bischof, T. Brox, & J.-M. Frahm (Eds.), Computer Vision—ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part I (pp. 162–177). Springer.
[6] Newton, D. (2019). Deep generative learning for the generation and analysis of architectural plans with small datasets. Proceedings of 37th eCAADe and 23rd SIGraDi Conference, (2), 21–28.
[7] Chen, M., Zheng, H., & Wu, J. (2022). Computational design of multi-functional system based on generative adversarial networks: Taking the layout generation of Vocational and Technical College as an example. Architectural Journal, (S1), 103–108.
[8] Lin, W. (2020). Research on automatic generation of primary school schoolyard layout based on deep learning [Master’s thesis]. South China University of Technology.
[9] Sun, C., Cong, X., & Han, Y. (2021). Generative design method of forced layout in residential area based on CGAN. Journal of Harbin Institute of Technology, 53(2), 111–121.
[10] Zhang, T. (2020). Experiments on generation of the arrangement of residential groups based on deep learning [Master’s thesis]. Nanjing University.
[11] Zhou, H., & Liu, H. (2021). Artificial intelligence aided design: Landscape plan recognition and rendering based on deep learning. Chinese Landscape Architecture, 37(1), 56–61.
[12] Qu, G., & Xue, B. (2022). Generative design method of landscape functional layout in residential areas based on Conditional Generative Adversarial Nets. Low Temperature Architecture Technology, 44(12), 5–9, 14.
[13] Chen, R., & Zhao, J. (2023). Generation and design feature recognition of landscape architecture scheme based on style-based generative adversarial network. Landscape Architecture, 30(7), 12–21.
[14] Zhao, G. (2023). Research on application of generative model in landscape design [Master’s thesis]. Shanxi University.
[15] Zhou, W. (2023). Design research on pocket park plan layout generation based on deep learning [Master’s thesis]. Chongqing Jiaotong University.
[16] Huang, Y., & Zhou, Y. (2023). Exploration on the generative architecture design method with AIGC technology: A case of the overall design process of generating architectural image with prompt as a key word. Urbanism and Architecture, 20(15), 202–206, 213.
[17] Chen, J., Shao, Z., & Hu, B. (2023). Generating interior design from text: A new diffusion model-based method for efficient creative design. Buildings, 13(7), 1861.
[18] Turrin, M., von Buelow, P., & Stouffs, R. (2011). Design explorations of performance driven geometry in architectural design using parametric modeling and genetic algorithms. Advanced Engineering Informatics, 25(4), 656–675.
[19] Isola, P., Zhu, J.-Y., Zhou, T., & Efros, A. A. (2017). Image-to-image Translation With Conditional Adversarial Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1125–1134). IEEE.
[20] Zhu, J.-Y., Zhang, R., Pathak, D., Darrell, T., Efros, A. A., Wang, O., & Shechtman, E. (2017). Toward Multimodal Image-to-image Translation. Proceedings of the 31st International Conference on Neural Information Processing Systems (pp. 465–476). Curran Associates Inc.
[21] Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017). Unpaired Image-to-image Translation Using Cycle-Consistent Adversarial Networks. Proceedings of the IEEE International Conference on Computer Vision (pp. 2223–2232). IEEE.
[22] Cha, S.-H., & Srihari, S. N. (2002). On measuring the distance between histograms. Pattern Recognition, 35(6), 1355–1370.
[23] Hore, A., & Ziou, D. (2010). Image Quality Metrics: PSNR vs. SSIM. 2010 20th International Conference on Pattern Recognition (pp. 2366–2369). IEEE.
[24] Geman, D., Geman, S., Hallonquist, N., & Younes, L. (2015). Visual turing test for computer vision systems. Proceedings of the National Academy of Sciences, 112(12), 3618–3623.
[25] Zhu, Y. (2022). Disordering and redirecting: Paradigm of design thinking in contemporary landscape architecture. World Architecture, (11), 36–37.
[26] Jiang, F., Ma, J., Webster, C. J., Li, X., & Gan, V. J. (2023). Building layout generation using site-embedded GAN model. Automation in Construction, (151), 104888.
[27] Li, P., Liu, B., & Gao, Y. (2018). An evidence-based methodology for landscape design. Landscape Architecture Frontiers, 6(5), 92–101.
[28] Yang, Y., & Lin, G. (2020). The development, connotations, and interests of research on landscape performance evaluation for evidence-based design. Landscape Architecture Frontiers, 8(2), 74–83.
[29] Zhou, H., Jiang, H., & Liu, H. (2021). Process visualization and performance evaluation of stormwater management in landscape projects based on IoT online monitoring. Chinese Landscape Architecture, 35(10), 29–34.
[30] Zhou, H., & Liu, H. (2021). IoT-based operational information management for built landscape projects: From vacancy to approaches. Landscape Architecture Frontiers, 9(2), 83–95.
[31] Zhou, H., Li, R., Liu, H., & Ni, G. (2023). Real-time control enhanced blue-green infrastructure towards torrential events: A smart predictive solution. Urban Climate, (49), 101439.
[32] Li, H., Zhang, Z., Liu, K., Chen, W., Wei, W., Liu, X., Xie, J., Zhang, M., Huang, Z., Zhong, M., Cai, C., Huang, X., Hou, Y., Lin, X., Yu, S., Fang, Y., & Feng, X. (2023, November 25). Toward dynamic optimization: Combining AI and EBHDL for the elderly. American Society of Landscape Architects.
[33] Chen, C., Li, H., Hou, Y., & Liu, J. (2023). Application progress of computer vision in the research on relationship between landscape and health. Landscape Architecture, 30(01), 30–37.
[34] Liu, H., Jin, C., & Yang, Y. (2023). Study on the programming language and its organicity of architectural generative design. Urbanism and Architecture, 20(5), 182–186.
System proposal of Globally Important Agricultural Heritage System.
[13] Jin, L. (2012). The theoretic roots of ecomuseums: The birth, development and practice of ecomuseums in China. Ecological Economy, (9), 180–185.
[14] Lin, X. (2020). Holistic protection: Values, concepts, practices, and challenges—Some thoughts on the innovation of cultural heritage conservation. Fujian Tribune (Humanities and Social Sciences Edition), (12), 36–47.
[15] Wu, X., & Wu, Q. (2021). From community presentation to panoramic integration: The transformation of the perspective of China’s eco-museum construction. Journal of North Minzu University (Philosophy and Social Science), (5), 16–24.
[16] Borrelli, N., & Davis, P. (2012). How culture shapes nature: Reflections on ecomuseum practices. Nature and Culture, 7(1), 31–47.
[17] Yin, K. (2017). Multiple-senses of place: An approach to ecomuseum. Folklore Studies, (5), 21–28, 158.
[18] Davis, P., & Li, M. (2019). Places, cultural exhibition sites and ecomuseums. Chinese Museum, (1), 36–42.
[19] Ohara, K. (2005). Ecomuseums in Japan today. Chinese Museum, (3), 58–62.
[20] Wang, M., Chen, Z., & Zhou, H. (2021). A feasibility study on the construction of eco-museum in the villages of She people in Eastern Fujian Province—A case study of the village of Banyueli in Xiapu County. Fujian Wenbo, (9), 39–44.
[21] Min, Q., He, L., & Sun, Y. (2009). Museum construction is an important content of agricultural cultural heritage protection: Summary of the Symposium on Agricultural Cultural Heritage Protection and Rural Museum Construction. Ancient and Modern Agriculture, (4), 114–116.
[22] Meng, F., Su, D., Fang, L., & An, L. (2017). Construction of ecology museum and development of ethnic culture: A case study of Suojia Ecology Museum. Journal of Original Ecological National Culture, 9(4), 128–140.
[23] Architectural History Institute of China Architecture Design & Research Group., & China Urban Development Planning & Design Consulting. (2022). Protection plan of Nanxun mulberry-dyke & fish-pond system of Huzhou City by cultural relics protection units (2022–2035) (Draft). People’s Government of Nanxun District, Huzhou City, Zhejiang Province.
[24] Zhejiang University. (2013). Protection and development plan of Nanxun mulberry-dyke & fish-pond system in Huzhou of Zhejiang Province. People’s Government of Nanxun District, Huzhou City, Zhejiang Province.
[25] Gu, X., Lou, L., Liu, M., & Min, Q. (2018). Review and prospect of studies on the dyke-pond system. Journal of Natural Resources, 33(4), 709–720.
[26] Shen, W. (2019). A research into the mulberry-based fish pond system on protective inheritance and innovational development in modern times—Taking Yunhao Family Farm in Nanxun, Huzhou, Zhejiang as an example. Research on Heritages and Preservation, 4(1), 34–38.
[27] Justice Bureau of Huzhou City. (2023). Announcement on public solicitation of opinions for Huzhou Mulberry-dyke & Fish-pond System protection regulations (Draft for review).
[28] Ren, Y., Li, C., & Zhang, S. (2023). The conservation of the mulberry fish pond as a cultural relic. Heritage, (1), 183–194.
[29] Qin, H. (2017). Narrative interpretation of the cultural value of canal heritage—Taking Tongzhou Canal in Beijing as an example. Journal of Beijing Union University (Humanities and Social Sciences), 15(4), 22–27.
[30] Li, X. (2019). Inheritance and practice of spatial pattern of traditional settlement landscape based on regional landscape in the south of Taihu Basin—Taking the landscape planning design of Nanxun-Qidu as an example [Master’s thesis]. Beijing Forestry University.
[31] China Academy of Urban Planning and Design., & Cultural Relics and Archaeology Institute of Zhejiang Province. (2011). Heritage conservation plan for the Zhejiang section of the Grand Canal.
[32] Chinese Academy of Agricultural Sciences., & Taihu Region Agricultural History Research Group of Chinese Agricultural Heritage Research Office of Nanjing Agricultural University. (1990). Draft of the Agricultural History of the Taihu Lake Region (pp. 181–182, 422). China Agriculture Press.
[33] Wang, C., & Zhang, R. (2020). A study on the interpretation system of cultural heritage: A case study of the Great Wall of Ming Dynasty in Beijing. Journal of Capital Normal University (Social Sciences Edition), (1), 139–149.
[34] ICOMOS. (2017). Charter for the interpretation and presentation of cultural heritage sites.
[35] National Cultural Heritage Administration of China. (2013). Nomination document for the Grand Canal of China as a World Heritage Site.
[36] Yin, K. (2019). Ecomuseum: Thought, Theory and Practice (p. 64). Science Press.
[37] Peng, Z., Qin, H., Guo Z., Zhou, K., Huang, H., Sun, M., Zhang, R., Li, Y., Liang, J., & Zhou, X. (2023). Interpretation and presentation: The contemporary construction and expression of multiple values of cultural heritage. Chinese Cultural Heritage, (3), 4–28.
[38] Qin, H. (2018). The interpretation path of the architectural heritage’s cultural value from the perspective of narration (II). Huazhong Architecture, 36(4), 1–3.
[39] Qin, H. (2018). The interpretation path of the architectural heritage’s cultural value from the perspective of narration (I). Huazhong Architecture, 36(3), 19–22.
[40] Chen, B., Qiu, Z., Usio, N., & Nakamura, K. (2018). Tourism’s impacts on rural livelihood in the sustainability of an aging community in Japan. Sustainability, 10(8), 2896.
[41] Davis, P., Gong, S., & Mai, X. (Trans.). (2023). Ecomuseums: A Sense of Place. Science Press.
本文引用格式 / PLEASE CITE THIS ARTICLE AS
Zhou, H., & Xiang, S. (2024). Applicability evaluation and reflection on Artificial Intelligence-based "image to image" generation of landscape architecture masterplans. Landscape Architecture Frontiers, 12(2), 58?67. https://doi.org/10.15302/J-LAF-1-020094
編輯 | 高雨婷,王穎
翻譯 | 高雨婷
制作 | 郭陽,高雨婷
媒體發布 | 馬哲
版權聲明:本文版權歸原作者所有,請勿以景觀中國編輯版本轉載。如有侵犯您的權益請及時聯系,我們將第一時間刪除。
投稿郵箱:info@landscape.cn
項目咨詢:18510568018(微信同號)