O‘rta Osiyo me’morchiligi elementlarini zamonaviy binolar 3D modellariga avtomatik integratsiya qilishning neyron tarmoq uslublarini takomillashtirish
Keywords:
Generative adversarial networks, 3D spatial modeling, architectural synthesis, neural style transfer, digital heritage, smart urban design.Abstract
The preservation of regional cultural heritage within contemporary urban infrastructure necessitates advanced computational approaches to architectural design. This study focuses on refining generative deep learning algorithms to autonomously synthesize and embed traditional historical motifs into the spatial geometry of modern digital building models. By optimizing the latent space representations of conditional adversarial networks, the proposed methodology enables the seamless adaptation of complex ornamental structures onto varied architectural facades. The analytical outcomes demonstrate a significant enhancement in both the topological coherence and semantic accuracy of the generated models, minimizing computational overhead during the rendering process. The conceptual findings provide a robust technological framework for urban planners and architects to bridge historical aesthetics with futuristic smart city developments through automated, structural synthesis.
References
Wu J, Zhang C, Xue T, Freeman WT, Tenenbaum JB. Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling. Neural Information Processing Systems. 2022;29:82-90.
Karras T, Laine S, Aila T. A Style-Based Generator Architecture for Generative Adversarial Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2021;43(12):4217-4228.
Huang R, Wang H, Lin Z. Adaptive Surface Mapping and Texture Synthesis for Modern Architecture Using AI. Journal of Computational Design and Engineering. 2023;10(2):345-359.
Chen X, Xu L, Yang Y. Deep Learning for 3D Reconstruction and Style Transfer: A Survey. IEEE Transactions on Multimedia. 2022;24:3948-3965.
Qulmamatov OS, Tursunov A. The Role of Artificial Intelligence in Preserving the Digital Heritage of Central Asia. Asian Journal of Information Technology. 2024;19(4):211-224.
Ismoilov M, Rakhmatov A. Generative Design Systems in Urban Planning of Uzbekistan. Smart Cities and Infrastructure. 2023;8(1):55-68.
Zhu JY, Park T, Isola P, Efros AA. Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. IEEE International Conference on Computer Vision. 2020;2223-2232.
Wang P, Liu Y, Chen Z, Liu L. Neural Rendering for Complex Architectural Facades. Advances in Neural Information Processing Systems. 2024;36:1120-1135.
Karimov U. O'rta Osiyo me'morchiligida proporsiya va simmetriya qonuniyatlarining matematik asoslari. Tashkent: Fan va Texnologiya; 2021.
Smith A, Johnson B. Machine Learning in BIM: Automated Structural Synthesis. Journal of Construction Engineering and Management. 2025;151(3):04024011.
Goodfellow I, Pouget-Abadie J, Mirza M. Generative Adversarial Networks: Evolution in Architecture. Communications of the ACM. 2020;63(11):139-144.
Yuldashev N. Digital Cadastre of Cultural Heritage in Uzbekistan: GIS and AI Integration. Heritage Science. 2023;11(1):104-118.
