臺灣博碩士論文加值系統

在電腦圖學中，貼圖技術被廣泛且普遍的運用在3D電腦遊戲或是虛擬實境的場景合成。材質合成演算法可建構出任意解析度的材質圖，成為貼圖技術中所需之大解析度材質圖，故乃成為一個重要的研究主題。
本篇論文提出一個新的材質合成演算法，我們稱為可操控的材質合成演算法。此演算法以區塊為合成的基本單位，能夠產生多樣化材質紋理特徵的合成結果。首先，我們導入控制場的概念，並根據材質的方向、大小、方位等三種特徵，產生相對應的控制場。其次，我們針對這三種材質特徵，做相對應的處理，產生合成時所需要的樣本材質。我們並將這些樣本材質建構成為KD-Tree樹狀架構，以增加材質合成時，需要的區塊搜尋、比對之效率。我們適當的調整參數，使的此樹狀架構在來源材質代表性與搜尋效率性間尋得一個平衡點。最後，我們進行可操控材質合成，即可輸出一張可經由使用者操控，具有方向、大小、方位特徵紋理變化的輸出材質。
我們實作所提的演算法，並且使用相關文獻上的材質作測試。實驗結果顯示，我們演算法所合成的輸出材質，展現使用者操控的材質紋理特徵。相較其他學者僅具方向、或僅具有方向、大小的演算法，我們的演算法更周延考慮兼具方向、大小、方位等材質紋理特徵，更能產生具有多樣化視覺效果的材質合成結果。在時間考量上，我們以區塊合成為基礎，並分析KD-Tree樹狀架構之建構參數，故能在維持一定的材質合成視覺品質下，有效的提升多樣性材質合成的速度。
總結本研究，可操控的材質合成演算法，同時考慮方向、大小、方位等三個材質特徵，故產生具有豐富變化、多樣性材質特徵的結果。兼具視覺品質與速度的考量，促使演算法能快速的產生材質合成圖。我們的演算法使的材質圖可以實際使用在貼圖技巧，增加場景合成的真實性，擴展電腦圖學的應用領域。

The texture mapping technique has been applied widely in computer graphics community to increase the visual realism of the 3D scenes. Texture synthesis algorithms construct an output texture with an arbitrary resolution suitable to be applied in the texture mapping technique. As a consequence, the study of texture synthesis algorithms is an important research topic.
In this thesis, we present a novel texture synthesis algorithm, the manipulative texture synthesis algorithm. The algorithm is based on the patch-based approach, and is able to generate various texture synthesis results. Our algorithm consists of three steps. First, we introduce the concept of the controllable field to manipulate the texture synthesis process. In particular, given an input texture, we analyze its features in terms of the direction, scale, and orientation, and then we generate three controllable fields corresponding to these features. Second, we generate a number of sample textures from the given input texture. Based on these sample textures, we construct a corresponding KD-Tree, allowing us to efficiently search the best matched sample texture for manipulative texture synthesis. During the sample texture generation, we also tune the parameter used to construct the nodes in the KD-Tree to make a balance between the quality of the synthesized result and the efficiency of the synthesis process. Finally, we execute our algorithm, which searches the best matched patch over the KD-Tree to produce a synthesized result.
Experiment results show that our algorithm is able to synthesize textures containing direction, scale, and orientation features. In addition, the synthesized texture has visually plausible appearance that is similar to the input texture. The result is produced within a few seconds, due to sophisticated KD-Tree generation, which balances the texture quality and the texture generation.
In conclusion, we propose a novel manipulative texture synthesis algorithm. To the best of our knowledge, we are the first to manipulate texture features in the direction, scale and the orientation, while our counterparts considered at most the features of direction and scale. Experimental results verify the feasibility of our algorithm in synthesizing textures with various features similar to the given input texture. Our algorithm can be applied to texture mapping to increase the realism of 3D scenes.

第一章 簡介 1
1.1 研究動機與目的 1
1.2 論文架構 3
第二章 相關文獻之探討 5
2.1 材質合成演算法之回顧 5
2.1.1 參數材質模型 6
2.1.2 像素材質合成演算法 6
2.1.3 區塊材質合成演算法 8
2.2 多樣性材質合成之相關研究 12
2.2.1 二維方向性材質合成 12
2.2.2 使用者定義材質合成 17
2.3 區塊材質合成的做法及流程 19
2.3.1 資料集建立部分 20
2.3.2 搜尋比對部分 23
2.3.3 貼圖部分 24
2.4 研究分析與心得 25
第三章 MTS演算法 26
3.1 產生控制場 27
3.1.1 方向控制場 27
3.1.2 縮放控制場 29
3.1.3 方位控制場 30
3.2 產生合成樣本材質 31
3.2.1 產生樣本 31
3.2.2 建立KD-Tree 33
3.3 可操控材質合成 36
3.3.1 搜尋比對樣本材質 38
3.3.2 搜尋比對材質區塊 40
3.3.3 區塊貼圖 43
第四章 實驗結果與分析 46
4.1 測試環境與測試模型 46
4.2 與傳統合成結果的各種比較與分析 48
4.3 由不同控制場所合成之多樣性結果展示. 53
4.3.1 單一材質與各種控制場的合成結果 53
4.3.2 不同材質與單一控制場的合成結果 54
4.4 與方向性材質合成相關文獻之比較與探討 61
4.4.1 MTS與Wang的比較 61
4.4.2 比較Taponecco的結果 62
4.5 MTS演算法的限制分析 62
第五章 結論與未來工作 64
5.1 結論 64
5.2 未來工作 65

[ANDR1999]G. Andrew and K. B. Eom, “Color Texture Synthesis with 2-D Moving Average Model,” Proceedings of 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 20-69, May 15-19, Phoenix, Arizona, 1999.
[ASHI2001]M. Ashikhmin, “Synthesizing Natural Texture,” Proceedings of ACM Symposium on Interactive 3D Graphics, pp. 217-226, March 2001.
[BONE1997]J. S. D. Bonet, “Multiresolution Sampling Procedure for Analysis and Synthesis of Texture Images,” Proceedings of SIGGRAPH 1997, pp. 361-368, August 1997.
[COHE2003]M. F. Cohen, J. Shade, S. Hiller, and O. Deussen, “Wang Tiles for Image and Texture Generation,” ACM Transactions on Graphics, Vol. 22, No. 3, pp. 287-294, July 2003.
[DISC2002]J. M. Dischler, K. Maritaud, B. Levy, and D. Ghazanfarpour, “Texture Particles,” Computer Graphics Forum, Vol. 21, No. 3, pp. 401-410, September 2002.
[DONG2005]F. Dong, H. Lin, and G. Clapworthy, “Cutting and Pasting Irregularly Shaped Patches for Texture Synthesis,” Computer Graphics Forum, Vol. 24, No. 1, pp. 17-26, 2005.
[EFRO1999]A. A. Efros and T. Leung, “Texture Synthesis by Non-Parametric Sampling,” Proceeding of International Conference on Computer Vision (ICCV ’99), Vol. 2, pp. 1033-1038, September 1999.
[EFRO2001]A. A. Efros and W. T. Freeman, “Image Quilting for Texture Synthesis and Transfer,” Proceedings of SIGGRAPH 2001, pp. 341-346, August 2001.
[FEDK2001]R. Fedkiw, J. Stam, and H. W. Jensen, “Visual simulation of smoke,” Proceedings of SIGGRAPH 2001, pp. 15-22, August 2001.
[GONG2003]Y. S. Gong, A Study of Effective Texture Synthesis by Edge Sampling, Master’s Thesis, Department of Computer Science, National Chung-Hsing University, Taiwan, 2003.
[GONZ2002]R. C. Gonzalez and R. E. Woods, Digital Image Processing, Second Edition, Prentice Hall, 2002.
[HEEG1995]D. J. Heeger and J. R. Bergen, “Pyramid-Based Texture Analysis /Synthesis,” Proceedings of SIGGRAPH 1995, pp. 229-238, 1995.
[HELM1991]J. L. Helm and L. Hesselink, “Visualizing Vector Field Topology in Fluid Flows,” IEEE Computer Graphics and Applications, Vol. 11, No. 3, pp. 36-46, May 1991
[JOLL1986]I. T. Jollife, Principal Components Analysis, Springer-Verlag, New York, 1986.
[KWAT2003]V. Kwatra, A. Schödl, I. Essa, G. Turk, and A. Bobick, “Image and Video Synthesis Using Graph Cuts,” ACM Transactions on Graphics (SIGGRAPH 2003), Vol. 22, No. 3, pp. 277-286, 2003.
[KWAT2005]V. Kwatra, I. Essa, A. Bobick, and N. Kwatra, “Texture Optimization for Example-based Synthesis,” ACM Transactions on Graphics (SIGGRAPH 2005), Vol. 24, No. 3, pp. 795-802, 2005.
[LEFE2005]S. Lefebvre and H. Hoppe, “Parallel Controllable Texture Synthesis,” ACM Transactions on Graphics (SIGGRAPH 2005), Vol. 24, No. 3, pp. 777-786, 2005.
[LIAN2001]L. Liang, C. Liu, Y. Q. Xu, B. Guo, and H. Y. Shum, “Real-Time Texture Synthesis by Patch-Based Sampling,” Proceedings of SIGGRAPH 2001, pp. 127-150, 2001.
[LIU2004]Y. Liu, W. C. Lin, and J. Hays, “Near-Regular Texture Analysis and Manipulation,” ACM Transactions on Graphics (SIGGRAPH 2004), Vol. 23, No. 3, pp. 368-376, 2004.
[MATU2005]W. Matusik, M. Zwicker, and F. Durand, “Texture Design Using a Simplicial Complex of Morphable Textures,” ACM Transactions on Graphics (SIGGRAPH 2005), Vol. 24, No. 3, pp. 787-794, 2005.
[MOUN1998]D. M. Mount, ANN Programming Manual, Technical Report, Department of Computer Science, University of Maryland, College Park, Maryland, USA, 1998.
[NEAL2003]A. Nealen and M. Alexa, “Hybrid Texture Synthesis,” Proceedings of Eurographics Symposium on Rendering 2003, pp. 97-105, 2003.
[NICO2005]A. Nicoll, J. Meseth, G. Müller, and R. Klein, “Fractional Fourier Texture Masks: Guiding Near-Regular Texture Synthesis,” Computer Graphics Forum, Vol. 24, No. 3, pp. 569-579, 2005.
[PORT1999]J. Portilla and E. P. Simoncelli, “Texture Modeling and Synthesis Using Joint Statistics of Complex Wavelet Coefficients,” in Proceedings of the IEEE Workshop on Statistical and Computational Theories of Vision, pp. 73-80, 1999.
[PORT2000]J. Portilla and E. P. Simoncelli, “A Parametric Texture Model Based on Joint Statistics of Complex Wavelet Coefficients,” International Journal of Computer Vision, Vol. 40, No. 1, pp. 49-70, October 2000.
[STAH2003]O. Stahlhut, “Extending Natural Textures with Multi-scale Synthesis,” Vision, Video, and Graphics, Vol. 67, No. 6, pp. 221-230, 2003.
[STAH2005]O. Stahlhut, “Extending Natural Textures with Multi-scale Synthesis,” Graphical Models, Vol. 67, No. 6, pp. 496-517, 2005.
[TAPO2004]F. Taponecco, “User-defined Texture Synthesis,” WSCG 2004, pp. 251-258, February 2004.
[TARI2005]M. Tarini1 and P. Cignoni, “Pinchmaps: Textures with Customizable Discontinuities,” Computer Graphics Forum, Vol. 24, No. 3, pp. 557-568, 2005.
[TURK2001]G. Turk, “Texture Synthesis on Surfaces,” Proceedings of ACM SIGGRAPH 2001, pp. 347-354, August 2001.
[WANG2004]B. Wang, W. Wang, H. Yang, and J. Sun, “Efficient Example-Based Painting and Synthesis of 2D Directional Texture,” IEEE Transactions on Visualization and Computer Graphics, Vol. 10, No. 3, pp. 266-277, 2004.
[WEI2000]L. Y. Wei and M. Levoy, “Fast Texture Synthesis Using Tree-Structured Vector Quantization,” Proceedings of SIGGRAPH 2000, pp. 479-488, July 2000.
[WU2004]Q. Wu and Y. Yu, “Feature Matching and Deformation for Texture Synthesis,” ACM Transactions on Graphics, Vol. 23, No. 3, pp. 362-365, August 2004.
[XUYQ2000]Y. Q. Xu, B. Guo, and H. Y. Shum, Chaos Mosaic: Fast and Memory Efficient Texture Synthesis, Microsoft Research Technical Reports, MsR-TR-2000-32, April 2000.
[ZELI2002]S. Zelinka and M. Garland, “Towards Real-Time Texture Synthesis with the Jump Map,” Proceedings of Eurographics Workshop on Rendering (EGWR 2002), pp. 20-27, 2002.
[ZELI2004]S. Zelinka and M. Garland, “Jump Map-Based Interactive Texture Synthesis,” ACM Transactions on Graphics, Vol. 23, No. 4, pp. 930-962, October 2004.