The Water Effect on Paper

When we apply water on paper, the wet area looks darker than the dry area. This phen

Figure 4.15 Color mixing. (a) The real image. (b) The simulated result.

et to the same value. Since we divide the concentration of ink into five levels in our proposed system, the reflectance of each level are 0.01, 0.1, 0.2, 0.3 and 0.4, respectively.

omenon is caused by the reflection and scattering of light on the paper surface. In our proposed system, we simulate the water effect in a simple way and define the wet area as a

mask. When users choose pigment type water to draw an area, the diffusion process works the same as that for ink and color pigment. Once the diffusion process is over, the number of carbon particles in each papel is use to construct an alpha mask; that is, the alpha value of each papel is in proportion to the number of the remaining carbon particles. We use this mask to filter the canvas and create an effect of water flowing into paper, as shown in Figure 3.4 in Chapter 3.

Chapter 5 Implementation Results

The implementation results of our proposed system are presented in this chapter. The system is written in C++ language and runs OpenGL on the PC platform with an AMD 1.4GHz CPU and 256 MB RAM.

Figure 5.1 shows the results of using our proposed system as Huang [13] did. Figure 5.1 (a) is the simulating result of dripping a drop of ink on Hsuan paper, which is done in interactive rate (within one second). And the effect of blending two brush, “dense brush following dilute brush” (濃墨破淡墨) process, is presented in Figure 5.1 (b). It is done by applying a dense brush stroke on the area where you have previously drawn a dilute brush stroke. And the dense stroke must be drawn before the dilute stroke is dried. Figure 5.1 (c) displays the stroke of dry brush by decreasing the water quantity of the brush (set the water number to be under 50).

Figure 5.1 Basic effects in Chinese Ink Painting. (a) A drop of ink. (b) Dense brush following dilute brush. (c) Strokes of dry brush.

Figure 5.2 (a) shows the blending result of several strokes with different concentrations.

By accumulating several strokes from thin to thick, users can get their desired ink blending effects in Chinese Ink Painting. Figure 5.3 (b) shows the blending effect of color pigments and ink.

Figure 5.2 (a) The sequence of several strokes. (b) The blending effect

In Figure 5.3 and Figure 5.4, we mimic two Chinese ink paintings drawn by Chang Dai-chien with the tablet pen device in our proposed system and the original paintings are attached for comparison. The painting time for the simulated images in Figure 5.3 and Figure 5.4 varies with the complexity of the original paintings: Figure 5.3 is within three hours while Figure 5.4 is about five hours. During the painting time, the ink diffusion process of each stroke is done in interactive rate (in seconds; the actual time depends on the size of the stroke:

the larger the size of the stroke, the longer the processing time).

Figure 5.3 (a) The original painting “Bamboo and Chrysanthemum” by Chang Dai-chien. (b) The simulated result.

Figure 5.4 (a) The original painting “Fisherman Return” by Chang Dai-chien. (b) The simulated result.

Figure 5.5 references a bird photo to draw an original Chinese Ink Painting with our proposed system. More original paintings drawn with the proposed system can be seen in Figure 5.6 to include more kinds of materials in Chinese Ink Painting.

Figure 5.5 Top: The painting using the proposed system. The photo in the right is used as a reference.

Figure 5.6 Several paintings using the proposed system.

How the Proposed System Works

Here we present a demonstration of mimicking a Chinese Ink Painting “Ling-chih Plants, Symbols Of Longevity” by Chang Dai-chien [7] to show how the proposed system works and

how to use the proposed system. We first depict the shape of the plants by selecting the brush type “normal brush” and pigment type “ink.” Note that we would like to use dry brush effect to draw the plants, so we don’t need high water quantity in the brush. The quantity of water is decreased to 52, as shown in Figure 5.7.

Figure 5.7 Using Dry Brush to Draw the Plants and Grass.

Next, we draw the leaves of the plants on the left by selecting the brush type “contour

Figure 5.8 Using Contour Brush to Draw the Plants on the Left.

Then we go on drawing the rock using normal brush. By changing the ink concentration, we can depict the texture of the rock. After the shapes of all objects in the painting are drawn, we then apply color on each object. First, apply color on the rock by selecting pigment type

“color pigment." After it is set, input the value of each RGB color channel and increase the quantity of water on brush to create more diffusion effects. Finally, we can draw the region of the rock by setting the stroke size to 6, as shown in Figure 5.9.

In Figure 5.10, we continue to apply color on the plant on the left. Since the leaves of the plant are thin, we don’t want the stroke to be diffused widely. The quantity of the water is decreased and another color is set to paint the leaves.

Figure 5.9 Apply color on the rock.

Figure 5.10 Apply color on the leaves of the plant.

Figure 5.11 shows the comparison of the final painted result (right) with the original painting (left).

Figure 5.11 Left: The original painting “Ling-chih Plants, Symbols Of Longevity” by Chang Dai-chien.

Right: The simulated result.

Chapter 6

Conclusions and Future Works

In this thesis, we propose an interactive system for Chinese color ink painting. The proposed system is composed of three main parts: a simple brush model to simulate Chinese writing brush, a method to generate ink diffusion effects in Chinese Ink Painting and a model to represent colors of pigments.

The brush model we proposed can produce good Chinese Painting strokes with tablet pen as the input device. By detecting the input pressure of the tablet pen, we can get various strokes with different shapes.

The diffusion model is based on real physical phenomenon. The paper mesh is separated into multi-layer X-Y planes, each of them divided into paper cells. Water particles flow in the holes or spaces of fibers of the paper mesh by capillarity. The carbon particles float and move in this liquid due to the collision with water particles. The directions and quantity of water flow are determined as a result of different degree of water absorbency, the alignment of fibers, and the factor of inertia of each paper cell. This ink diffusion algorithm can be used to draw lots of subjects with ink diffusion effect in Chinese Ink Painting style by controlling the user-defined stroke parameters. The most important is the effect of expressing the mixture of

other. “Dense brush following dilute brush”, is a typical example in Chinese Ink Painting.

Besides, since the proposed algorithms are based on physical theory and observational-based analyses, resulting images using this physical-based method are very realistic.

By adapting KM theory, the color of each pigment is correctly computed. First, we transform the user input RGB reflectance into reflection and transmission coefficients. Then we apply KM equations with the transformed coefficients to get the reflectance R in each RGB color channel on paper. The color of several overlaying pigments can also be represented accurately by computing the reflectance R in each RGB color channel with the reflection and the transmission coefficients.

Further works have to be done:

1. Although the brush model we proposed can generate fair stroke results, there are still some effects cannot be generated: the brush splitting effect and the various brush effects created by deforming the brush hair and changing the hand gesture while painting. In the future, we hope to improve our brush model by proposing a more complex one to simulate all the effects of Chinese Ink Painting.

2. The processing time of the ink diffusion process depends on the area of the input stroke.

The larger the input stroke area is, the longer time the diffusion process takes. Although we can achieve interactive rate during painting, the input strokes with large area still take lots of time to finish the diffusion process. Therefore, we will implement the ink diffusion process on hardware to decrease the processing time of the proposed system.

3. The resolution of our proposed system is restricted (the current resolution is 600*600 pixels). We refresh the whole screen pixels in each time step even when the stroke only

occupies few parts of the whole screen. For the efficiency of the refreshing time, the screen resolution should not be too large. As to this problem, we will try to do some improvements in the future, such as subdivide the whole screen into several parts and only refresh the parts where the stroke occupies.

4. Currently, the color representation of our proposed system simulates the composition of color pigments. Besides the effects of overlaying color pigment, there are still some other common mixing effects created by mixing colors on canvas in real paintings. We will attach the mixing color on canvas feature to our proposed system in the future to create various mixing color effects on canvas.

Reference

[8] Bill Baxter, Vincent Scheib, Ming C. Lin, Dinesh Manocha, “DAB: Interactive Haptic Painting with 3D Virtual Brushes”, Proceedings of ACM SIGGRAPH 01, 2001, pp.

461-468.

[9] Cassidy J. Curtis, Sean E. Anderson, Jonathon. E. Seims, and Kurt W. Fleischer, and David H. Salesin, “Computer-generated watercolor”, Proceedings of SIGGRAPH 97, in Computer Graphics Proceedings, Annual Conference Series, pp. 421-430, August 1997.

[10] Colin Doncaster, “Toon - A shader for cartoon rendering” (for RenderMan) http://www.webnation.com/vidirep/panels/renderman/shaders/toon/toon.phtml

[11] Qinglian Guo, Tosiyasu L. Kunii, “Modeling the Diffuse Paintings of ‘Sumie’”. In: Kunii T (Ed.), Modeling in Computer Graphics (Proceedings of the IFIP WG5.10). Berlin:

Springer, 1991. Page 329~338.

[12] Chet S. Haase, Gary W. Meyer, “Modeling Pigmented Materials for Realistic Image Synthesis”, ACM Transactions on Graphics (TOG), Volume 11, Issue 4 (October 1992),

1992, Pages: 305 - 335.

[13] Sheng-Wen Huang, Der-Lor Way, Zen-Chung Shih, “Physical-Based Model of Ink Diffusion in Chinese Ink Paintings”, Journal of WSCG 2003.

[14] Henrik Wann Jensen, Justin Legakis, Julie Dorsey, “Rendering of Wet Materials”, Rendering Techniques '99. Eds. D. Lischinski and G. W. Larson. Springer-Verlag, pages

273-282, 1999.

[15] G Kortum, “Reflectance Spectroscopy”, Springer-Verlag, 1969.

[16] Jintae Lee, “Diffusion rendering of black ink paintings using new paper and ink models”, Computer & Graphics 25 (2001), Page 295~308.

[17] Jintae Lee, “Simulating oriental black-ink painting”, IEEE Computer Graphics &

Applications 1999; 19(3): 74-81.

[18] Suguru Saito and Masayuki Nakajima, “3D physics-based brush model for painting", Proceedings of ACM SIGGRAPH 99:conference abstracts and applications, Page 226

August 8 - 13, 1999, Los Angeles, CA USA.

[19] David L Small, “Simulating Watercolor by Modeling Diffusion , Pigment and Paper Fibers”, SPIE Proceedings, Vol. 1460, No. 15. San Jose, CA, 1990.

[20] Steve Strassmann, ”Hairy Brushes”, Proceedings of ACM SIGGRAPH 86, page 225-232.

[21] Theo G. M., Van De Ven, “Colloidal Hydrodynamics”, Colloid Science – A Series of Monographs, Page 51~77.

[22] Der-Lor Way, Chih-Wei Hsu, Hsin-Yi Chiu, Zen-Chung Shih, “Computer-Generated Chinese Painting for Landscapes and Portraits”, Journal of WSCG 2001.

[23] Der-Lor Way, Yu-Ru Lin, Zen-Chung Shih, “The Synthesis of Trees in Chinese Landscape Painting Using Silhouette and Texture Strokes”, Journal of WSCG Volume 10, Number 3, 2002, pp. 449-507.

[24] Der-Lor Way, Zen-Chung Shih, “The Synthesis of Rock Textures in Chinese Landscape

[25] Shan-Zan Wen, Zen-Chung Shih, Hsin-Yi Chiu, “The Synthesis of Chinese Ink Painting”, National Computing Symposium’99, page 461-468, 1999.

[26] Andrew Witkin, “Particle system Dynamics”, Physically Based Modeling (SIGGRAPH 2001 COURSE NOTES), Page C1~C12.

[27] H.T.F. Wong, H.H.S. Ip, “Virtual brush: a model-based synthesis of Chinese calligraphy”, Computers and Graphics, Volume 24, Number, 2000, pp. 99-113.

[28] Songhua Xu, Francis C.M. Lau, Feng Tang, Yunhe Pan, “Advanced Design for a Realistic Virtual Brush”, Computer Graphics Forum Volume22, Issue 3, 2003, pp.

533-542(10).

[29] Young Jung Yu, Do Hoon Lee, “Interactive Rendering Technique for Realistic Oriental Painting”, Journal of WSCG, Volume11, Number1, 2003.