Peyzaj Mimarlığında Metinden Görüntüye: Yapay Zekâ Yaklaşımları

Abstract

Peyzaj mimarlığı tasarım çalışmalarında, yaratıcı ve farklı konseptlere sahip mekânların yaratılmasında fikir geliştirme süreci, uzun zamanlar gerektiren bir süreçtir. Teknolojik gelişmelerle birlikte zamanın etkin kullanımı önem kazanmakta ve yapay zekâ (AI) alanında yaratıcı, gerçekçi ve detaylı metinden görsel üreten uygulamalar dikkat çekmektedir. Text-to-Image (T2I) yöntemi ile tasarım fikirlerinin hızlı ve doğru bir şekilde geliştirilmesi açısından, peyzaj mimarlığında AI uygulamalarının kullanımı, akıllara birçok soru getirmektedir. AI tarafından üretilen görseller, mesleki doğruluk ve estetik açıdan ne kadar tatmin edicidir? mesleki terimlerini doğru bir şekilde algılayabilmekte midir? AI ve peyzaj mimarlığı arasındaki genel ilişki durumu nedir? gibi sorulara bu çalışmada cevap aranmaktadır. Bu çalışmada; tasarım alanında kullanılan 50 mesleki terim belirlenmiş ve bu terimler doğrultusunda 6 farklı tasarım konsepti ile 70 kelimelik promptlar oluşturulmuştur. Oluşturulan promptlar kullanılarak, T2I özelliğine sahip Dall-E, MidJourney, LookX ve mnml uygulamalarından görseller üretilmiştir. Üretilen görseller; mesleki uygunluk, görsel estetik, yaratıcılık ve teknik detaylar açısından değerlendirilmiştir. Çalışma sonucunda, seçilen terimler arasında bazı mesleki terimlerin soyut, ileri teknik ve kapsam alanın geniş kavramlar olduğu tespit edilmiştir. AI uygulamalarında mesleki terimlerin doğru bir şekilde algılanabilmekte, peyzaj tasarımında farklı ve yaratıcı konseptlerin geliştirilmesinde tatmin edici görseller elde edebilmektedir. Bu bağlamda, peyzaj mimarlığı ve AI alanında iş birliği yapılarak uygulamaların geliştirilmesine katkı sağlanabilir. Aynı zamanda, bu çalışma; T2I uygulamaları ve peyzaj mimarlığı arasındaki mevcut durumu ortaya koyarak, daha yaratıcı, kaliteli ve detaylı konsept görsellerin üretilmesi açısından geliştirilebilir bir potansiyel olduğunu göstermektedir.

Keywords

• Yapay Zeka
• Metinden Görsele
• Peyzaj Mimarlığı
• Peyzaj Tasarımı
• Konsept

Text to Image in Landscape Architecture: Artificial Intelligence Approaches

Abstract

Generating ideas for designing spaces with creative and diverse concepts in landscape architecture design is a time-consuming process. With technological advancements, effective use of time has become increasingly significant, and applications that produce creative, realistic, and elaborate visuals from text in the field of artificial intelligence (AI) have attracted attention. However, the use of AI applications in landscape architecture raises many questions regarding the rapid and accurate development of design ideas through the Text-to-Image (T2I) method: how satisfying are AI-generated visuals in terms of professional accuracy and aesthetics? Can AI accurately perceive professional terms? What is the general relationship between AI and landscape architecture? Apart from coming up with an answer to these questions, this study aims to investigate the potential of text-to-image (T2I) models in generating design concepts for landscape architecture. Fifty professional terms were selected and used to create 70-word prompts across six distinct design concepts. Four popular T2I models (Dall-E, MidJourney, LookX, and mnml) were employed to generate visuals based on these prompts. The images generated were evaluated based on their adherence to professional standards, aesthetic appeal, creativity, and technical accuracy. Results indicated that while AI models could effectively interpret a wide range of professional terms, including abstract and highly technical concepts, there were limitations in capturing the nuanced details of landscape design. This study highlights the potential of AI to assist landscape architects in the early stages of the design process, but also underscores the need for human expertise to refine and optimize AI-generated designs. Future research should explore ways to improve the accuracy and specificity of AI-generated landscape designs, as well as investigate the potential of integrating AI with other design tools and techniques.

Keywords

• Artificial Intelligence
• Text-to-Image
• Landscape Architecture
• Landscape Design
• Concept

References

1. Agnese, J., Herrera, J., Tao, H. & Zhu, X. (2019). A survey and Taxonomy of Adversarial Neural Networks for Text-to-image Synthesis. WIREs Data Mining and Knowledge Discovery. https://doi.org/10.1002/widm.1345.
2. Anonymous (2023, August). Mimarların ve Tasarımcıların Bilmesi Gereken 10 Yapay Zeka Uygulaması. XXI. https://xxi.com.tr/i/mimarlarin-ve-tasarimcilarin-bilmesi-gereken-10-yapay-zeka-uygulamasi.
3. Ardhianto, P., Santosa, Y. P., Moniaga, C., Utami, M. P., Dewi, C., Christanto, H. J., Chen, A. P. S. (2023). Generative Deep Learning for Visual Animation in Landscape Design. Hindawi Scientific Programming. https://doi.org/10.1155/2023/9443704
4. Aslan, T. & Aydın, K. (2023). Metinden Görüntü Üretme Potansiyeli Olan Yapay Zekâ Sistemleri Sanat ve Tasarım Performanslarının İncelenmesi. Ondokuz Mayıs University Journal of Faculty of Education, 42(2), 1149-1198. https://doi.org/10.7822/omuefd.1293657.
5. Benliay, A. & Kılıç, A. (2024). Peyzaj Tasarımı Sunum Tekniklerinde Yapay Zeka Uygulamalarının Değerlendirilmesi. PEYZAJ – Eğitim, Bilim, Kültür ve Sanat Dergisi. 6(1), 1-14. DOI: 10.53784/peyzaj.1490265
6. Bengesi, S., El-Sayed, H., Sarker, M., Houkpati, Y., Irungu, J., & Oladunni, T. (2023). Advancements in Generative AI: A Comprehensive Review of GANs, GPT, Autoencoders, Diffusion Model, and Transformers. IEEE Access, 12, 69812-69837.
7. Chen, B., Zhang, Z., Langrene, N., & Zhu, S. (2023). Unleashing the Potential of Prompt Engineering in Large Language Models: A Comprehensive Review. ArXiv.
8. Dang, H., Mecke, L., Lehmann, F., Goller, S., & Buschek, D. (2022). How to Prompt? Opportunities and Challenges of Zero- and Few-Shot Learning for Human-AI Interaction in Creative Applications of Generative Models. ArXiv.

9. Das, S., Dey, A., Pal, A. & Roy, N. (2015). Applications of Artificial Intelligence in Machine Learning: Review and Prospect. International Journal of Computer Applications, 115(9), 31-41.
10. Denerel, S. B. & Birişçi, T. (2019). A Research on Landscape Architecture Student Use of Traditional and Computer-aided Drawing Tools. Amazonia Investiga, 8(24), 373-385.
11. Enholm, I. D., Papagiannidis, E., Mikalef, P. & Krogstie, J. (2022). Artificial Intelligence and Business Value: a Literature Review. Information Systems Frontiers. 24, 1709-1734. https://doi.org/10.1007/s10796-021-10186-w
12. Enjellina, Beyan, E. V. P. & Rossy A. G. C. (2023). A Review of AI Image Generator: Influences, Challenges, and Future Prospects for Architectural Field. Journal of Artificial Intelligence in Architecture, 2(1), 53–65. https://doi.org/10.24002/jarina.v2i1.6662
13. Fernandez, P. (2022). Technology Behind Text to Image Generators. Library Hi Tech News, 10, 1-4. http://dx.doi.org/10.1108/LHTN-10-2022-0116
14. Fernberg, P., George, B. H. & Chamberlain, B. (2023). Producing 2D Asset Libraries with AI-powered Image Generators. Journal of Digital Landscape Architecture, 8, 186-194. 10.14627/537740020
15. Frolov, S., Hinz, T., Raue, F., Hees, J. & Dengel, A. (2021). Adversarial Text-to-Image Synthesis: A Review. Neural Networks, 144, 187 – 209. https://doi.org/10.1016/j.neunet.2021.07.019
16. Gafni, O., Polyak, A., Ashual, O., Sheynin, S., Parikh, D., & Taigman, Y. (2022). Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors. ArXiv.
17. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. & Bengio, Y. (2020). Generative Adversarial Networks. Communications of the ACM, 63(11), 139-144. http://dx.doi.org/10.1145/3422622
18. Graff, D. (2023, 6 November). “mnml.ai: Redefining Architectural Design with AI-Powered Innovation”. Artificial Intelligencepedia. https://artificialintelligencepedia.com/mnml-ai-redefining-architectural-design-with-ai-powered-innovation/.
19. Gui, J., Sun, Z., Wen, Y., Tao, D. & Ye, J. (2023). A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications. IEEE Transactions on Knowledge and Data Engineering, 35(4), 3313- 3332. 10.1109/TKDE.2021.3130191
20. Hanafy, N. O. (2023). Artificial Intelligence’s Effects on Design Process Creavitiy: “A Study on Used A.I. Text-to-Image in Architecture”. Journal of Building Engineering, 80, 107999. https://doi.org/10.1016/j.jobe.2023.107999
21. Hong, Y., Hwang, U., Yoo, J., & Yoon, S. (2017). How Generative Adversarial Networks and Their Variants Work. ACM Computing Surveys (CSUR), 52, 1 - 43.
22. Howard, J. (2019). Artificial Intelligence: Implications for the Future of Work. American Journal of Industrial Medicine, 62: 917-926. 10.1002/ajim.23037
23. Jiang, Y., Li, X., Luo, H., Yin, S. & Kaynak, O. (2022). Quo Vadis Artificial Intelligence?. Discover Artificial Intelligence, 2, 4. https://doi.org/10.1007/s44163-022-00022-8
24. Li, M. (2023). Designer Robots: An Early Look at Applications for Artificial Intelligence Visualization Software in Landscape Architecture. Master Thesis. The University of Guelph, Canada.
25. Li, M. & Amoroso, N. (2023). An Early Look at Applications for Artificial Intelligence Visualization Software in Landscape Architecture. Journal of Digital Landscape Architecture, 543 – 553. 10.14627/537740057.
26. Lin, Z. (2024). How to write Effective Prompts for Large Language Models. Nature Human Behaviour 2024(8), 611–615. https://doi.org/10.1038/s41562-024-01847-2.
27. Lo, L. S. (2023). The Art and Science of Prompt Engineering: A New Literacy in the Information Age. Internet Reference Services Quarterly, 27(4), 203–210. https://doi.org/10.1080/10875301.2023.2227621
28. LookX (2024, 18 June). About Us. URL: https://www.lookx.ai/about.html
29. Lund, B. (2023). The Prompt Engineering Librarian. Library Hi Tech News, 8(40), 6-8. https://doi.org/10.1108/LHTN-10-2023-0189.
30. Lyu, Y., Wang, X., Lin, R. & Wu, J. (2022). Communication in Human–AI Co-Creation: Perceptual Analysis of Paintings Generated by Text-to-Image System. Applied Sciences, 12(22), 11312. https://doi.org/10.3390/app122211312.
31. Makarouni, E. (2024, 18 June). Tech for Architects: 7 Top AI Tools for Architectural Rendering and Visualization. Architizer. https://architizer.com/blog/practice/tools/top-ai-tools-for-architectural-rendering-visualization/.
32. Minh, D., Wang, H. X., Li, Y. F. & Nguyen, T. N. (2022). Explainable Artificial Intelligence: A Comprehensive Review. Artificial Intelligence Review, (55), 3503-3568. https://doi.org/10.1007/s10462-021-10088-y.
33. Mijwel, M. M. (2015). History of Artificial Intelligence. Computer Science,(April 2015), 3-5.
34. Morandin-Ahuerma, F. (2022). What is Artificial Intelligence? International Journal of Research Publication and Reviews, 3(12), 1947 – 1951.
35. Oppenlaender, J. (2023). A Taxonomy of Prompt Modifiers for Text-to-Image Generation. Behaviour & Information Technology, 1–14. https://doi.org/10.1080/0144929X.2023.2286532
36. Pouya, S. (2020). Technical English for Students of Landscape Architecture. Malatya: İnönü University Publisher.
37. Sağlık, A. & Minkara, E. B. (2024). Perception in Human Pyschology and Landscape Prioritised Visualisation of Space with Artificial Intelligence. Journal of Architectural Sciences and Applications, 9(2), 831 – 843. https://doi.org/10.30785/mbud.xxxxx
38. Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E.L., Ghasemipour, S.K., Ayan, B.K., Mahdavi, S.S., Lopes, R.G., Salimans, T., Ho, J., Fleet, D.J., & Norouzi, M. (2022). Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding. ArXiv.
39. Sahoo, P., Sahoo, P., Singh, A. K., Saha, S., Jain, V., Mondal, S., & Chadha, A. (2024). A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications. ArXiv.
40. Sharma, N., Sharma, R. & Jindal, N. (2021). Machine Learning and Deep Learning Appliacations-A Vision. Global Transitions Proceedings, 2, 24-28. https://doi.org/10.1016/j.gltp.2021.01.004
41. Tan, Y.X., Lee, C.P., Neo, M., Lim, K.M., Lim, J.Y., & Alqahtani, A. (2023). Recent Advances in Text-to-Image Synthesis: Approaches, Datasets and Future Research Prospects. IEEE Access, 11, 88099-88115.
42. Tanugraha, S. (2023). A Review Using Artificial Intelligence-Generating Images: Exploring Material Ideas from MidJourney to Improve Vernacular Designs. JARINA – Journal of Artificial Intelligence in Architecture, 2(1), 48 – 57.
43. Tecuci, G. (2012). Artificial Intelligence. WIREs Comput Stat, 2012; 4: 168-180. 10.1002/wics.200
44. Wang, J., Liu, Z., Zhao, L., Wu, Z., Ma, C., Yu, S., Dai, H., Yang, Q., Liu, Y., Zhang, S., Shi, E., Pan, Y., Zhang, T., Zhu, D., Li, X., Jiang, X., Ge, B., Yuan, Y., Shen, D., Liu, T. & Zhang, S. (2023). Review of large vision models and visual prompt engineering. Meta-Radiology, 2023(1), 100047. https://doi.org/10.1016/j.metrad.2023.100047.
45. Waterman, T. (2009). The Fundamentals of Landscape Architecture. AVA Academia the Environment of Learning.
46. White, J., Quchen, F., Hays, S., Sandborn, M., Olea, C., Gilbert, H., Elnashar, A., Spencer-Smith, J. & Schmidt, D. C. (2023). A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT. ArXiv. https://doi.org/10.48550/arXiv.2302.11382.
47. Żelaszczyk, M., & Ma'ndziuk, J. (2024). Text-to-Image Cross-Modal Generation: A Systematic Review. ArXiv.
48. Zhang, H., Xu, T., Li, H., Zhang, S., Wang, X., Huang, X., & Metaxas, D.N. (2017). StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks. 2017 IEEE International Conference on Computer Vision (ICCV), 5908-5916.
49. Zhu, J., Park, T., Isola, P., & Efros, A.A. (2017). Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks. 2017 IEEE International Conference on Computer Vision (ICCV), 2242-2251.