Katmanlı 2B İllüstrasyonlardan Fulldome Sahne Kurulumu için Karma-İnisiyatifli Prosedürel Bir İş Akışı

Öz

Katmanlı 2B illüstrasyonların fulldome ortamlarına aktarımında iki temel güçlük vardır: manuel 3B modelleme yüzlerce sahne öğesi içeren sahnelere ölçeklenmez; yapay zekâ tabanlı monoküler derinlik tahmini yöntemleri ise stilize ve panoramik içerikte alan kayması nedeniyle nesne sınırlarını koruyamaz ve deterministik sanatçı kontrolü sağlayamaz. Mevcut yaklaşımlar prosedürel otomasyon, sanatçı güdümlü kalibrasyon ve dome'a özgü yerleşimi tek bir karma-inisiyatifli akışta birleştirmek yerine ayrı ayrı ele almaktadır. Bu çalışma, katmanlı 2B illüstrasyonları sabit dome kamerası altında yeniden hesap gerektirmeyen kararlı yerleşim ve yönelimle 3B fulldome sahnelere dönüştüren karma-inisiyatifli prosedürel bir iş akışı önermektedir. Sistem, ufuk parametreli bir derinlik haritalama stratejisiyle figürlerin resimsel yüksekliğini radyal derinliğe çevirir; 360° düzen için silindirik haritalama, zemin için örtük yüzey çıkarımına dayalı figür tabanlı arazi uyumu uygular. Hiyerarşik parametrik kalibrasyonla sanatçı, derinlik ve ölçeği hem global hem kategori düzeyinde denetleyebilir. İş akışı, 400 sahne öğesi içeren tarihsel bir canlandırma projesi üzerinde geliştirilip sınanmıştır. Sistem, uygun sahne öğelerinin tamamında prosedürel yerleşim sağlamış, öğe bazlı manuel düzeltmeye gerek kalmamış ve yalnızca üç parametre ayarlama iterasyonu yeterli olmuştur. Alfa-maskesi piksel kesişimiyle otomatik belirlenen 400 örtüşen figür çiftinde (manuel seçim olmadan) kaynak katman ilişkileriyle oklüzyon sıra uyumu 357/400 (%89,25); raycast sonrası ortalama zemin temas düzeltmesi 0,62 m (düzeltmelerin %78'i ±1 m aralığında) olarak ölçülmüştür. Üretim süresinin yaklaşık 30 kat azaldığı tahmin edilmektedir. Bu bulgular, kural tabanlı derinlik haritalama ile karma-inisiyatifli parametrik kalibrasyonun stilize panoramik içerikte fulldome sahne kurulumu için uygulanabilir, deterministik ve sanatçı kontrollü bir yaklaşım sunduğunu göstermektedir.

Anahtar Kelimeler

• prosedürel içerik üretimi
• fulldome görselleştirme
• karma-inisiyatifli tasarım
• derinlik haritalama sezgiseli
• sahne kurulumu
• dijital kültürel miras görselleştirme

A Mixed-Initiative Procedural Workflow for Fulldome Scene Assembly from Layered 2D Illustrations

Öz

Transferring layered 2D illustrations to immersive fulldome environments is challenging: manual 3D modeling does not scale to scenes containing several hundred elements, while AI-based monocular depth estimation (MDE) suffers from domain shift on stylized and panoramic content and lacks deterministic artist control. Existing approaches address procedural automation, artist-driven calibration, and dome-specific layout separately rather than in one mixed-initiative pipeline. This study proposes a mixed-initiative procedural workflow that transforms layered 2D illustrations into 3D fulldome scenes with stable placement and billboard orientation under a fixed dome camera, requiring no recomputation under camera rotation. The system uses a horizon-parameterized depth mapping heuristic converting pictorial height into radial depth, cylindrical mapping for 360° layout, and figure-derived terrain fitting via implicit surface reconstruction. Hierarchical parametric calibration preserves artist control over depth and scale at global and category levels. The workflow was tested on a 400-element historical visualization project. It achieved full procedural placement coverage across eligible scene elements with no per-element manual correction, completed in three parameter-adjustment iterations. Across 400 automatically identified overlapping figure pairs, determined by pixel-level alpha-mask intersection without manual selection, occlusion-order consistency with source-layer relations was 357/400 (89.25%). The mean ground-contact correction after raycasting was 0.62 m, with 78% of corrections within ±1 m. Production time was estimated to improve approximately 30-fold. Rule-based depth mapping with mixed-initiative calibration thus offers a viable, deterministic, artist-controllable approach for fulldome scene assembly from stylized panoramic content.

Anahtar Kelimeler

• procedural content generation
• fulldome visualization
• mixed-initiative design
• depth mapping heuristic
• scene assembly
• digital heritage visualization

Kaynakça

1. Bourke, P. (2005). Spherical mirror: A new approach to hemispherical dome projection. Planetarian, 34(4), 6–9.
2. Candy, L., & Edmonds, E. (2018). Practice-based research in the creative arts: Foundations and futures from the front line. Leonardo, 51(1), 63–69. https://doi.org/10.1162/LEON_a_01471
3. Cutting, J. E., & Vishton, P. M. (1995). Perceiving layout and knowing distances: The integration, relative potency, and contextual use of different information about depth. In W. Epstein & S. Rogers (Eds.), Perception of space and motion (pp. 69–117). Academic Press. https://doi.org/10.1016/B978-012240530-3/50005-5
4. Denard, H. (Ed.). (2009). The London Charter for the computer-based visualisation of cultural heritage (Version 2.1). https://www.londoncharter.org
5. Gardner, J. S., Austerweil, J. L., & Palmer, S. E. (2010). Vertical position as a cue to pictorial depth: Height in the picture plane versus distance to the horizon. Attention, Perception, & Psychophysics, 72(2), 445–453. https://doi.org/10.3758/APP.72.2.445
6. Horry, Y., Anjyo, K., & Arai, K. (1997). Tour into the picture: Using a spidery mesh interface to make animation from a single image. SIGGRAPH '97, 225–232. https://doi.org/10.1145/258734.258854
7. Karadağ, D. (2025). AI in architectural education: Rethinking studio culture. PLANARCH – Design and Planning Research, 9(2), 243–253. https://doi.org/10.54864/planarch.1749891
8. Kazhdan, M., Bolitho, M., & Hoppe, H. (2006). Poisson surface reconstruction. Eurographics Symposium on Geometry Processing, 61–70. https://doi.org/10.2312/SGP/SGP06/061-070

9. Kerbl, B., Kopanas, G., Leimkühler, T., & Drettakis, G. (2023). 3D Gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics, 42(4), Article 139. https://doi.org/10.1145/3592433
10. Koskinen, I., Zimmerman, J., Binder, T., Redström, J., & Wensveen, S. (2012). Design research through practice: From the lab, field, and showroom. Morgan Kaufmann. https://doi.org/10.1016/C2010-0-65896-2
11. Lantz, E. (2007). A survey of large-scale immersive displays. In Proceedings of the 2007 Workshop on Emerging Displays Technologies: Images and Beyond. ACM. https://doi.org/10.1145/1278240.1278241
12. Liapis, A., Yannakakis, G. N., & Togelius, J. (2013). Sentient Sketchbook: Computer-aided game level authoring. FDG '13, 213–220.
13. López-Menchero Bendicho, V. M., & Grande, A. (2011). International principles of virtual archaeology (the Seville Principles). Spanish Society of Virtual Archaeology (SEAV) / International Forum of Virtual Archaeology. (Ratified by the 19th ICOMOS General Assembly, New Delhi, 2017.)
14. Maruani, N., Zhang, P., Chaudhuri, S., Fisher, M., Zhao, N., Kim, V. G., Alliez, P., Desbrun, M., & Wang, Y. (2025). Illustrator's depth: Monocular layer index prediction for image decomposition. arXiv:2511.17454 [Preprint]. https://doi.org/10.48550/arXiv.2511.17454
15. Ranftl, R., Lasinger, K., Hafner, D., Schindler, K., & Koltun, V. (2022). Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(3), 1623–1637. https://doi.org/10.1109/TPAMI.2020.3019967
16. Renner, R. S., Velichkovsky, B. M., & Helmert, J. R. (2013). The perception of egocentric distances in virtual environments: A review. ACM Computing Surveys, 46(2), Article 23. https://doi.org/10.1145/2543581.2543590
17. Shade, J., Gortler, S., He, L., & Szeliski, R. (1998). Layered depth images. SIGGRAPH '98, 231–242. https://doi.org/10.1145/280814.280882
18. Shaker, N., Togelius, J., & Nelson, M. J. (2016). Procedural content generation in games. Springer. https://doi.org/10.1007/978-3-319-42716-4
19. Smith, G., Whitehead, J., & Mateas, M. (2010). Tanagra: A mixed-initiative level design tool. FDG '10, 209–216. https://doi.org/10.1145/1822348.1822376
20. Szeliski, R., & Shum, H.-Y. (1997). Creating full view panoramic image mosaics and environment maps. SIGGRAPH '97, 251–258. https://doi.org/10.1145/258734.258861
21. Szeliski, R., Anandan, P., & Baker, S. (1999). From 2D images to 2.5D sprites: A layered approach to modeling 3D scenes. IEEE International Conference on Multimedia Computing and Systems, 1, 44–50. https://doi.org/10.1109/MMCS.1999.779118
22. Szymanowicz, S., Insafutdinov, E., Zheng, C., Campbell, D., Henriques, J. F., Rupprecht, C., & Vedaldi, A. (2024). Flash3D: Feed-forward generalisable 3D scene reconstruction from a single image. arXiv:2406.04343 [Preprint]. https://doi.org/10.48550/arXiv.2406.04343
23. Talton, J. O., Gibson, D., Yang, L., Hanrahan, P., & Koltun, V. (2009). Exploratory modeling with collaborative design spaces. ACM Transactions on Graphics, 28(5), Article 167. https://doi.org/10.1145/1618452.1618513
24. van Dijk, T., & de Croon, G. (2019). How do neural networks see depth in single images? Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2183–2191. https://doi.org/10.1109/ICCV.2019.00227
25. Yang, L., Kang, B., Huang, Z., Xu, X., Feng, J., & Zhao, H. (2024). Depth anything: Unleashing the power of large-scale unlabeled data. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 10371–10381. https://doi.org/10.1109/CVPR52733.2024.00987
26. Young, M. (2018). Drawn to 360°: How can the aesthetics and qualities of traditional 2D animation storytelling add to the immersive VR projection paradigm? Animation Practice, Process & Production, 7(1), 11–40. https://doi.org/10.1386/ap3.7.1.11_1
27. Yu, H.-X., Duan, H., Herrmann, C., Freeman, W. T., & Wu, J. (2025). WonderWorld: Interactive 3D scene generation from a single image. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 5916–5926. https://doi.org/10.1109/CVPR52734.2025.00555
28. Zimmerman, J., Forlizzi, J., & Evenson, S. (2007). Research through design as a method for interaction design research in HCI. CHI '07, 493–502. https://doi.org/10.1145/1240624.1240704