Raqamli Transformatsiya va Sun’iy Intellekt
3D MODELLASHTIRISHDA MATEMATIK FUNKSIYALAR ASOSIDA AVTOMATIK YARATILADIGAN MODELLAR, SPLAYNLAR VA SUN’IY INTELLEKT USULLARI ASOSIDA TEKSTURA YARATISH
1 Muhammad al-Xorazmiy nomidagi TATU
Zenodo DOI: 10.5281/zenodo.20717706
Keywords
Abstract
Ushbu maqolada 3D modellashtirish masalalari uchun zamonaviy tekstura yaratish usullari, jumladan matematik funksiyalar asosida avtomatik yaratiladigan modellar, splayn usullari, tasvirga asoslangan yondashuvlar hamda sun’iy intellektga asoslangan texnologiyalar o‘rganiladi. Tekstura generatsiyasining matematik asoslari keng qamrovda tahlil qilingan bo‘lib, ular orasida shovqin funksiyalari, fraktal modellar, splayn interpolatsiyasi, tasvirga asoslangan sintez usullari va neyron tarmoqlarga asoslangan algoritmik arxitekturalar mavjud.
Alohida e’tibor tekstura usullarini obyektlarning geometrik ifodasi bilan integratsiyalashga qaratilgan, ayniqsa splayn egri chiziqlari asosida qurilgan modellar bilan bog‘lash orqali 3D obyekt yuzasida tekstura joylashuvining aniqligi va mosligi oshiriladi.
Maqolada asosan matematik algoritmlar asosida avtomatik obyektlar yoki teksturalarni yaratish usullari, jumladan Perlin shovqini va fraktal funksiyalar, patch-based yondashuvga asoslangan tasvir sintezi usullari hamda GAN va CNN arxitekturalariga asoslangan zamonaviy neyron tarmoq modellari ko‘rib chiqilgan.
Mazkur ishda avtomatik hosil qilinadigan, tasvirga asoslangan va sun’iy intellekt usullarini birlashtiruvchi gibrid tekstura generatsiya modeli taklif etiladi. Ushbu yondashuv hisoblash samaradorligi, fotorealistik tasvir sifati va moslashuvchanlik o‘rtasida optimal muvozanatni ta’minlaydi.
Usullar vizual sifat, hisoblash murakkabligi va xotira sarfi bo‘yicha taqqoslanib tahlil qilingan.
Tajriba natijalari shuni ko‘rsatdiki, gibrid yondashuv alohida usullarga nisbatan yuqori vizual sifatni ta’minlaydi hamda faqat neyron tarmoqlarga asoslangan yechimlarga qaraganda hisoblash xarajatlarini kamaytiradi. Ushbu natijalar zamonaviy grafik tizimlarda, jumladan VR/AR muhitlari va raqamli obyektlarni rekonstruksiya qilishda tekstura generatsiyasi uchun turli yondashuvlarni integratsiya qilishning yuqori salohiyatga ega ekanligini tasdiqlaydi.
References
- Gatys, L.A., Ecker, A.S., Bethge, M.: Texture synthesis using convolutional neural networks. arXiv preprint arXiv:1505.07376 (2015).
- Goodfellow, I. et al.: Algoritmike adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2672–2680 (2014).
- Karras, T. et al.: Analyzing and improving the image quality of StyleGAN. In: Proceedings of CVPR (2020).
- Liu, X. et al.: Neural texture synthesis and style transfer. arXiv preprint arXiv:1705.06830 (2017).
- Mallat, S.: Group invariant scattering. Communications on Pure and Applied Mathematics (2012).
- Cook, R.L.: Stochastic sampling in computer graphics. ACM Transactions on Graphics 5(1), 51–72 (1986).
- Li, Y., Wand, M.: Combining Markov random fields and convolutional neural networks for image synthesis. In: Proceedings of CVPR (2016).
- Huang, J., Wu, C.: Hybrid procedural and neural texture generation for real-time rendering. Computer Graphics Forum (2023).
- Szeliski, R.: Computer Vision: Algorithms and Applications. Springer, London (2010).
- Efros, A.A., Freeman, W.T.: Image quilting for texture synthesis and transfer. In: Proceedings of SIGGRAPH, pp. 341–346 (2001).
- Ebert, D.S., Musgrave, F.K., Peachey, D., Perlin, K., Worley, S.: Texturing and Modeling: A Procedural Approach, 3rd edn. Morgan Kaufmann, San Francisco (2003).
- Coughlan, J.M., Yuille, A.L.: Image-based rendering using texture mapping. IEEE Transactions on Pattern Analysis and Machine Intelligence 24(11), 1501–1515 (2002).
- Mandelbrot, B.B.: The Fractal Geometry of Nature. W.H. Freeman, New York (2004).
- Hertzmann, A. et al.: Image analogies. In: Proceedings of SIGGRAPH (2001).
- Guo, J. et al.: Deep learning for texture synthesis: A survey. Computer Graphics Forum (2021).
- Turk, G.: Texture synthesis. In: ACM SIGGRAPH Course Notes (2001).
- Elad, M., Milanfar, P.: Style-transfer via texture optimization. IEEE Transactions on Image Processing (2019).
- Foydalanilgan adabiyotlar
- Peitgen, H.-O., Jürgens, H., Saupe, D.: Chaos and Fractals: New Frontiers of Science. Springer, New York (2006).
- Akenine-Möller, T., Haines, E., Hoffman, N.: Real-Time Rendering, 4th edn. CRC Press, Boca Raton (2018).
- Zhou, Y. et al.: Texture modeling with deep neural networks. IEEE Access (2020).
- Shirley, P., Marschner, S.: Fundamentals of Computer Graphics, 4th edn. CRC Press, Boca Raton (2015).
- Johnson, J., Alahi, A., Fei-Fei, L.: Perceptual losses for real-time style transfer and super-resolution. In: European Conference on Computer Vision (ECCV) (2016).
- Kingma, D.P., Welling, M.: Auto-encoding variational Bayes. arXiv preprint arXiv:1312.6114 (2013).
- Lefebvre, S., Hoppe, H.: Appearance-space texture synthesis. ACM Transactions on Graphics 25(3), 541–548 (2006).
- Musgrave, F.K.: Fractal terrains and procedural textures. In: SIGGRAPH Course Notes (1993).
- Lagae, A., Lefebvre, S., Drettakis, G., Dutré, P.: Procedural noise using sparse Gabor convolution. ACM Transactions on Graphics 28(3), 1–10 (2009).
- Karras, T. et al.: Progressive growing of GANs for improved quality, stability, and variation. In: International Conference on Learning Representations (ICLR) (2018).
- Zhang, H. et al.: Deep texture synthesis with CNNs. IEEE Transactions on Visualization and Computer Graphics 23(9), 2227–2239 (2017).
- Gatys, L.A. et al.: A neural algorithm of artistic style. arXiv preprint arXiv:1508.06576 (2015).
- Isola, M. et al.: Image-to-image translation with conditional GANs. In: Proceedings of CVPR (2017).