臺灣博碩士論文加值系統

在計算機圖學（computer graphics）的範疇中，中國水墨畫（Chinese in painting）的模擬是非擬真成圖（non-photorealistic rendering, NPR）領域�堶垠n的研究主題之一。本論文提出了兩個以影像為基礎（image-based）、非筆觸的（non-stroke）成圖演算法，來將一張輸入影像自動合成為一張具有彩色水墨擴散（color ink diffusion）、溼中溼水拓效果（wet-in-wet flow effect）、以及黑邊效果（dark-edge effect）－－三種常見於水墨畫效果的繪圖風格影像。首先，我們提出一個以影像為基礎的彩色水墨成圖演算法（IBCIDR）來模擬彩色水墨擴散風格。接著，在IBCIDR的基礎之上，我們提出另一個有效的以影像為基礎的彩色水墨擴散合成演算法（EIBCIDS）來產生更好的彩色水墨擴散、溼中溼水拓（wet-in-wet flow effect）、以及黑邊效果。
以影像為基礎的彩色水墨成圖演算法（IBCIDR）包含三個主要的階段：特徵提取階段、混色階段、以及彩色水墨擴散合成階段。在特徵提取階段，利用亮度分段及色彩分割（luminance division and color segmentation, LDCS）來將參考影像的資訊抽象化；在混色階段，庫貝卡－蒙克理論（Kubelka-Munk theory, K-M theory）用來計算兩種顏料重疊時的混色結果；接著，在彩色水墨擴散合成階段，我們提出一個以自然法則為基礎的（physically-based）彩色水墨擴散模型，來模擬彩色水墨滴在一張用材質合成（texture synthesis）技術所產生的吸水宣紙上的擴散現象。我們的IBCIDR演算法不僅改進了傳統水墨擴散模擬只侷限在黑白水墨的缺點，同時也展示了一種不需要建構任何筆觸，將彩色影像自動轉換成令人激賞的水墨擴散風格影像的方法。
IBCIDR是第一個非筆觸、以影像為基礎的自動水墨擴散風格成圖演算法。然而，美中不足的是它有四個缺點－－“色彩偏移”、“細節遺失”、無溼中溼水拓效果、無黑邊效果。因此，我們再提出一個新的、有效的以影像為基礎的彩色水墨擴散合成演算法（EIBCIDS）來克服這四個缺點。
EIBCIDS演算法包含三個主要部分：有效的彩色水墨擴散合成（ECIDS）、可調控的SK-M黑邊混色、可調控的溼中溼水拓技術（CWFT）。在ECIDS演算法中，我們提出一個全新更靈敏的庫貝卡－蒙克理論（SK-M theory）來進正確的顏料混色計算，也提出一個新的重疉公式（NOL）來正確重疊沉積層和擴散層，使得擴散影像得以呈現與原始參考影像更近似的色調和更翔實的細節。此外，我們也提出新的影像抽象法來產生更平滑的抽象結果。在可調控的SK-M黑邊混色演算法中，用邊界萃取法得到的黑邊資訊，和由ECIDS產生的擴散影像，將經由SK-M公式進行黑邊混色。在這過程中，使用者可以透過邊界門檻參數和邊界強度參數來進行調控。而可調控的溼中溼水拓技術被設計成獨立機制，以減低水墨擴散模擬的複雜度。我們使用可變長線積分迴旋（ALLIC）來表示參考影像的全域流場，並透過參考影像的亮度值和使用者調控的比重參數，來建立可調控的流場圖。再利用SK-M公式來將流場圖和擴散影像混合成具有溼中溼水拓效果的影像。
我們的EIBCIDS演算法具有視覺真實、可調控、獨立和簡單等四個優點。

Chinese ink painting simulation is an important research area of non-photorealistic rendering (NPR) in the computer graphics community. This thesis proposes two image-based, non-stroke, painterly rendering algorithms for automatically synthesizing an image with color ink diffusion, wet-in-wet flow effect, and dark-edge effect -- three famous effects existing in the Chinese ink painting. We first present an image-based color ink diffusion rendering (IBCIDR) algorithm for the simulation of color ink diffusion. Based on this IBCIDR, we next suggest an effective image-based color ink diffusion synthesis (EIBCIDS) algorithm to mimic better diffusion, wet-in-wet flow, and dark-edge effects.
The image-based color ink diffusion rendering (IBCIDR) algorithm contains three main phases. In the feature extraction phase, the information of the reference image is simplified by luminance division and color segmentation (LDCS). In the color mixing phase, the Kubelka-Munk (K-M) theory is employed to approximate the result when one pigment is painted upon another pigment layer. Then in the color ink diffusion synthesis phase, the physically-based color ink diffusion model (CIDM) that we propose is employed to simulate the result of color ink diffusion in absorbent paper using a texture synthesis technique. Our IBCIDR algorithm eliminates the drawback of conventional Chinese ink simulations, which are limited to the black ink domain, and our approach demonstrates that, without using any strokes, a color image can be automatically converted to the diffused ink style with a visually pleasing appearance.
The IBCIDR is the first, non-stroke, image-based color ink diffusion rendering algorithm. However, the IBCIDR also showed four drawbacks-- “color shifting,” “detail missing,” no wet-in-wet flow effect, and no dark-edge effect. Consequently, we propose a new, effective, non-stroke, image-based color ink diffusion synthesis algorithm (called EIBCIDS) to overcome these four problems.
The EIBCIDS contains three main stages: effective color ink diffusion synthesis (ECIDS), controllable SK-M edge blending, and controllable wet-in-wet flow technique (CWFT). In the ECIDS algorithm, we propose a new K-M equation (SK-M) to properly mix pigment color. We also present a new color ink diffusion synthesis algorithm which uses a new overlapping equation (NOL) to properly overlap the deposit layer with a diffusion layer. Thus, the diffused image can show more similar tone and more ink diffusion details than IBCIDR. In addition, a new image abstraction approach is suggested to generate a smoother abstraction from the reference image. In the controllable SK-M edge blending algorithm, the suitable edges are extracted from the reference image by edge extraction approach. By employing the SK-M equation, the edge map created by edge extraction is blended with the diffused image which is simulated by ECIDS. The user can control the detail and strength of dark-edge effect by using an edge threshold coefficient and an edge strength coefficient. The controllable wet-in-wet flow technique (CWFT) is suggested, independent of the ink diffusion algorithm to decrease the complexity of the entire simulation system. In particular, we use the adaptive length line integral convolution (ALLIC) to represent the global flow of the reference image. Referring to the luminance of the reference image and the desired weight coefficient provided by the user, a controllable flow map is constructed. Finally, we employ our SK-M equation again to blend the controllable flow map with the diffused image to generate the final rendered result. Our EIBCIDS algorithm has four advantages: visually plausible, controllable, independent, and simple.

Acknowledgments ........................................................................................................... i
中文摘要 ........................................................................................................................ ii
Abstract ......................................................................................................................... iv
1 Introduction ............................................................................................................ 1
2 Related work ...........................................................................................................7
2.1 Non image-based ............................................................................................. 7
2.2 Image-based .................................................................................................... 9
2.3 Other related works ....................................................................................... 10
3 Image-Based Color Ink Diffusion Rendering .................................................... 13
3.1 Physically-based color ink diffusion model .................................................. 13
3.1.1 Synthesizing paper texture ..................................................................... 16
3.1.2 Taking account of gravity ..................................................................... 18
3.1.3 Deposit layer and diffusion layer ........................................................... 18
3.2 Image-Based Color Ink Diffusion Rendering ............................................... 20
3.2.1 Feature extraction ................................................................................... 21
3.2.2 Rendering steps ...................................................................................... 25
3.3 Results ..................................................................................................... 29
3.4 Summary .................................................................................................... 41
4 Effective Image-Based Color Ink Diffusion Synthesis ..................................... 42
4.1 Effective color ink diffusion synthesis .......................................................... 43
4.1.1 Image abstraction ................................................................................... 44
4.1.2 Edge extraction ...................................................................................... 47
4.1.3 ECIDS overview .................................................................................... 49
4.1.4 A new sensitive Kubelka-Munk equation .............................................. 50
4.1.5 A new overlapping equation .................................................................. 52
4.2 Controllable wet-in-wet flow technique ....................................................... 53
4.2.1 Controllable flow map creation .............................................................. 53
4.2.2 SK-M flow blending .................................................................................... 56
4.3 Controllable dark-edge blending ........................................................................ 58
4.4 Results ........................................................................................................ 61
4.5 Summary ....................................................................................................... 69
5 Conclusions and future work .............................................................................. 70
5.1 Conclusion .................................................................................................... 70
5.2 Future work ....................................................................................................73
Bibliography .............................................................................................................. 74
Index .......................................................................................................................... 81

[BTS2006]A. Bousseau, M. Kaplan, J. Thollot, and F. X. Sillion, “Interactive watercolor rendering with temporal coherence and abstraction,” Proceedings of the 4th international Symposium on Non-Photorealistic Animation and Rendering (NPAR ''06), pp. 141-149, 2006.
[CA2002]C. Chan and E. Akleman, “Two Methods for Creating Chinese Painting,” Proceedings of 10th Pacific Conference on Computer Graphics and Applications (PG’02), pp. 403-412, 2002.
[CAS+1997]C. J. Curtis, S. E. Anderson, J. E. Seims, K. W. Fleischery, and D. H. Salesin, “Computer-Generated Watercolor,” Proceedings of 24th Annual Conference on Computer Graphics & Interactive Techniques, ACM SIGGRAPH, ACM Press, pp. 421-430, 1997.
[CL2006]M. T. Chi and T. Y. Lee, “Stylized and Abstract Painterly Rendering System Using a Multiscale Segmented Sphere Hierarchy,” IEEE Transactions on Visualization and Computer Graphics, Vol. 12, No. 1, pp. 61-72, 2006.
[CM2002]D. Comanicu and P. Meer, “Mean Shift: A Robust Approach toward Feature Space Analysis,” IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 24 pp. 603-619, 2002.
[CT2004]S. H. Chu and C. L. Tai, “Real-Time Painting with an Expressive Virtual Chinese Brush,” IEEE Computer Graphics and Applications, Vol. 24, No. 5, pp. 76-85, 2004.
[CT2005]N. S.-H. Chu and C.-L. Tai, “MoXi: Real-Time Ink Dispersion in Absorbent Paper,” ACM Transactions on Graphics (SIGGRAPH 2005 issue), Vol. 24, No. 3, pp. 504-511, August 2005.
[EF2001]A. Efros and W. T. Freeman, “Image Quilting for Texture Synthesis and Transfer,” Proceedings of SIGGRAPH 2001, pp. 341-346, August 2001.
[Farbiz2003]F. Farbiz, A. D. Cheok, and P. Lincoln, “Automatic Asian art: computers converting photos to Asian paintings using humanistic fuzzy logic rules,” Proceedings of the SIGGRAPH 2003, Sketches & Applications, p. 1, 2003.
[FK1999]M. Friedman and A. Kandel, Introduction to Pattern Recognition Statistical, Structural, Neural and Fuzzy Logic Approaches, Imperial College Press, London, 1999.
[GCS2002]B. Gooch, G. Coombe, and P. Shirley, “Artistic vision: painterly rendering using computer vision techniques,” Proceedings of the second international symposium on non-photorealistic animation and rendering, ACM Press, pp. 83-90, 2002.
[GK2003]Q. L. Guo. And T. L. Kunii, “Nijimi rendering algorithm for creating quality black ink paintings,” Proceedings of the Computer Graphics international (CGI’03), pp. 152-159, 2003.
[GW2002]R. C. Gonzalez and R. E. Woods, Digital Image Processing, 2nd edition, Prentice-Hall, 2002.
[HE2004]J. Hays, and I. Essa, “Image and Video Based Painterly Animation,” Proceedings of the 3rd international symposium on non-photorealistic animation and rendering, ACM Press, pp. 113-120, 2004.
[Her1998]Hertzmann, “Painterly Rendering with Curved Brush Strokes of Multiple Sizes,” Proceedings of ACM SIGGRAPH 98, pp. 453-460, 1998.
[HM1992]C. S. Haase and G. W. Meyer., “Modeling Pigmented Materials for Realistic Image Synthesis,” ACM Transactions on Graphics, Vol. 11, No. 4, pp. 305-335, 1992.
[HWC2003]S. W. Huang, D. L. Way, and Z. C. Shih, “Physical-based Model of Ink Diffusion in Chinese Paintings,” The 11th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG ''2003), pp. 520-527, 2003.
[KN2001]T. L. Kunii and G. V. Nosovskij, “Two-Dimensional Diffusion Model for Diffusion Ink Painting,” International Journal of Shape Modeling, Vol. 7, No. 1, pp. 45-58, 2001.
[KNH1995]T. L. Kunii, G. V. Nosovskij, and T. Hayashi, “A Diffusion Model for Computer Animation of Diffuse Ink Painting,” Proceedings of the Computer Animation, pp. 98-102, 1995.
[Kwo1990]D. W. Kwo, Chinese Brushwork in Calligraphy and Painting: Its History, Aesthetics, and Techniques, Dover Publications, Inc. New York, 1990.
[Lee1999]J. Lee, “Simulating Oriental Black-Ink Painting,” IEEE Computer Graphics and Applications, Vol. 19, No. 3, pp. 74-81, May/June 1999.
[Lee2001]J. Lee, “Diffusion Rendering of Black Ink Paintings Using New Paper and Ink Models,” Computers & Graphics, Vol. 25, pp. 295-308, 2001.
[LLT2004]S. Lee, H. Y. Lee, I. F. Lee and C. Y. Tseng, “Ink Diffusion in Water,” European Journal of Physics, Vol. 25, No. 2, pp. 331-336, 2004.
[PM1990]P. Perona and J. Malik, “Scale-space and edge detection using anisotropic diffusion, “ IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 12, No. 7, pp. 629-639, 1990.
[SH1995]D. H. Stalling and C. Hege, “Fast and resolution independent line integral convolution,” Proceedings of the ACM SIGGRAPH ’95 Conference on Computer Graphics, pp. 249-256, 1995.
[Sma1990]D. L. Small, “Simulating Watercolor by Modeling Diffusion, Pigment and Paper Fibers,” SPIE Proceedings, Vol. 1460, No. 15, San Jose, CA, 1990.
[TNF1999]S. Takagi, M. Nakajima and I. Fujishiro, “Volumetric Modeling of Colored Pencil Drawing,” Proceedings of the 7th Pacific Conference on Computer Graphics and Applications, pp. 250-258, 1999.
[WI2000]H. T. F. Wong and HHH. H. S. IpHH, “Virtual Brush: a Model-based Synthesis of Chinese Calligraphy,” Computer & Graphics, Vol. 24, pp. 99-113, 2000.
[WL2003]C. M. Wang, and J. S. Lee, “Using ILIC algorithm for an impressionist effect and stylized virtual environments,” International Journal of Visual Languages and Computing, Vol. 14, No. 3, pp. 255-274, 2003.
[WS2001]D. L. Way, and Z. C. Shih, “The Synthesis of Rock Textures in Chinese Landscape Painting,” Computer Graphics Forum, Vol. 20, No. 3, pp. C123-C131, 2001.
[WSC1999]S. Z. Wen, Z. C. Shih, and H. Y. Chiu, “The Synthesis of Chinese Ink Painting,“ Proceedings of National Computing Symposium ’99 (NCS’99), Taiwan, pp. 461-468, 1999.
[WW2007]C. M. Wang and R. J. Wang, ”Image-Based Color Ink Diffusion Rendering,” IEEE Transactions on Visualization and Computer Graphics, Vol. 13, No. 2, pp. 235-246, 2007.
[WW2007a]R. J. Wang and C. M. Wang, ”Image-based Color Ink Diffusion with Controllable Flow Effect,” Proceedings of the 20th International Conference on Computer Animation and Social Agents (CASA2007), pp. 55-62, Hasselt, Belgium, June 2007.
[WW2007b]R. J. Wang and C. M. Wang, ”Effective Color Ink Diffusion Synthesis,” To Appear in Proceedings of the 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIHMSP07), 2007.
[WWL2003]C. M. Wang, R. J. Wang, and J. S. Lee, “A Study of Physically-Based Ink Diffusion,” Journal of Technology, Vol. 18, No. 4, pp. 393-402, 2003. (in Chinese)
[WWS2006]D. L. Way, W. J. Lin, and Z. C. Shih, "Computer-Generated Chinese Color Ink Paintings," Journal of the Chinese Institute of Engineers, Vol. 29, No. 6, pp. 1041-1050, 2006.
[XTL+2002]S. H. Xu, M. Tang, F. Lau, and Y. H Pan, “A Solid Model Based Virtual Hairy Brush,” Proceedings of Eurographics 2002, pp299-308, 2002.
[XXK+2006]S. Xu, Y. Xu, S. B. Kang, D. H. Salesin, Y. Pan, and H. Y. Sum, “Animating Chinese Paintings Through Stroke-Based Decomposition,” ACM Transactions on Graphics, Vol. 25, No. 2, pp. 239-267, 2006.
[YLC2002]Y. J. Yu, Y. B. Lee, and H. G. Cho, “A Model Based Technique for Realistic Oriental Painting,” Proceedings of 10th Pacific Conference on Computer Graphics and Applications (PG’02), pp. 452-453, 2002.
[YLO2002]J. S. Yeh, T. Y. Lien, and M. Ouhyoung, “On the Effects of Haptic Display in Brush and Ink Simulation for Chinese Painting and Calligraphy,” Proceedings of 10th Pacific Conference on Computer Graphics and Applications (PG’02), pp. 439-441, 2002.
[YLP2003]J. H. Yu, G. M. Luo, and Q. S. Peng, “Image-based synthesis of Chinese landscape painting,” Journal of Computer Science and Technology, Vol. 18, No. 1, pp. 22-28, 2003.
[YO2002]J. W. Yeh and M. Ouhyoung, “Non-Photorealistic Rendering in Chinese Painting of Animals,” Journal of System Simulation, Vol. 14, No. 6, pp. 1220-1224, 2002.
[ZST+1999]Q. Zhang, Y. Sato, J. Y. Takahashi, K. Muraoka, and N. Chiba, “Simple Cellular Automaton-Based Simulation of Ink Behaviour and Its Application to Suibokuga-Like 3D Rendering of Trees,” The Journal of Visualization and Computer Animation, Vol. 10, pp. 27–37, 1999.