系統識別號 | U0002-1408201314472300 |
---|---|
DOI | 10.6846/TKU.2013.00389 |
論文名稱(中文) | 影像特效轉換系統 |
論文名稱(英文) | An image effect converter system |
第三語言論文名稱 | |
校院名稱 | 淡江大學 |
系所名稱(中文) | 資訊工程學系博士班 |
系所名稱(英文) | Department of Computer Science and Information Engineering |
外國學位學校名稱 | |
外國學位學院名稱 | |
外國學位研究所名稱 | |
學年度 | 101 |
學期 | 2 |
出版年 | 102 |
研究生(中文) | 白逸群 |
研究生(英文) | Yi-Chun Pai |
學號 | 897410162 |
學位類別 | 博士 |
語言別 | 英文 |
第二語言別 | |
口試日期 | 2013-07-04 |
論文頁數 | 41頁 |
口試委員 |
指導教授
-
林慧珍(086204@mail.tku.edu.tw)
委員 - 謝錦棠(hsieh@ee.tku.edu.tw) 委員 - 廖宏源(liao@iis.sinica.edu.tw) 委員 - 顏淑惠(105390@mail.tku.edu.tw) 委員 - 施國琛(timothykshih@gmail.com) 委員 - 林慧珍(086204@mail.tku.edu.tw) |
關鍵字(中) |
影像抽象化 雙向濾波器 高斯濾波器 卡通化 非寫實渲染 shock filter Kuwahara filter |
關鍵字(英) |
image abstraction bilateral filter Gaussian blur filter cartoon animation non-photorealistic rendering shock filter Kuwahara filter |
第三語言關鍵字 | |
學科別分類 | |
中文摘要 |
在本論文中,我們提出了一個能夠有效的影像特效轉換系統,並能夠滿足不同使用者的設計需求。為了完成這樣的系統,我們會先介紹基礎影像處理效果,接著說明本系統如何產生出各種不同的影像特效。在實驗結果的部分,展現出了許多不同影像特效的結果。 |
英文摘要 |
This dissertation proposes an image effect converter system that can create various types of effects for images, to satisfy different needs of users for specific design goals. Some operations needed for creating image effects are first proposed, and various image effects that can be converted by the proposed system are illustrated. The experiments show very rich results in terms of image effects. |
第三語言摘要 | |
論文目次 |
Table of Contents Table of Contents III List of Figures IV Chapter 1 Introduction 1 Chapter 2 Basic operators 4 2.1 Difference of Gaussian (DoG) 4 2.2 Kuwahara Filter 5 2.3 Bilateral Filter 5 2.4 Optional Color Quantization (OCQ) 6 2.5 Line segment filter 6 2.6 Tone adjustment 7 2.7 Shock Filter 9 2.8 Continuous Glass Pattern 10 Chapter 3 The proposed operations 13 3.1 Flow-based Bilateral Filter 13 3.2 Flow-based Gaussian filter 16 3.3 Curve-shaped filters 18 3.4 Line drawing 19 3.5 Pencil texture generator 21 3.6 Modified shock filter 22 Chapter 4 Proposed image effect creation methods 24 4.1 Image abstraction 24 4.2 Pencil sketch 26 4.3 Watercolor painting 27 4.4 Artistic imaging with Continuous Glass Pattern 28 Chapter 5 Experimental results 31 Chapter 6 Conclusion and future works 38 References 39 List of Figures Figure 1. Examples of the line segment filters with different radians (a). θ0 = 0, (b). θ1 = π/8, (c). θ2 = π/4, (d). θ4 = π/2. 7 Figure 2. Tone distributions in a natural image and in a pencil drawing [24]. 7 Figure 3. An example of tone adjustment (a). original image, (b). histogram, (c). distribution Bright, (d). distribution Mild, (e). distribution Dark, (f). combination of (c)~(e) with ratio 5:2:1, (g). result. 8 Figure 4. Examples of CESF [30] (a). original images, (b). CESF (σ = 2, t = 10), (c) CESF (σ = 4, t = 10). 10 Figure 5. An example of Glass pattern [6] (a). vector field ((y2-1)+(1/3)xy, (1/3)(y2-1)-xy)), (b). the trajectories solving the corresponding differential equation, (c). a corresponding GP. 10 Figure 6. Continuous Glass Pattern 11 Figure 7. An example of CGP (a). original image, (b). synthetic painterly texture, (c). resulting painterly CGP image. 12 Figure 8. Two versions of bilateral filter, where the red dots denote pixel x, and the blue lines/curves and black line/curve are the traced paths along gradient direction/flow and tangent direction/flow, respectively. (a) orientation-aligned bilateral filter, (b) flow-based bilateral filter 13 Figure 9. Comparison of OABF and FBBF (a). original image, (b). result of OABF, (c). result of FBBF, (d~f) close-ups of (a~c). 15 Figure 10. Neighbors collected by (a) directional Gaussian filter (b) flow-based Gaussian filter 16 Figure 11 (a). original image, (b). result of directional Gaussian filtering (c). result of flow-based Gaussian filtering 17 Figure 12. Comparison of traditional Gaussian filter and the flow-based Gaussian filter (a). original image (b). traditional Gaussian filter (σ = 2) (c). flow-based Gaussian filter (σg = 0.33, σt = 9) 17 Figure 13. Comparisons of the original Gaussian filter and the flow-based Gaussian filter under different parameter settings (a). original image, (b). result of traditional Gaussian filter with σ = 1, (c).~(h). results of flow-based Gaussian filter with (c). σg = 0.33, σt = 3, (d). σg = 1, σt = 3, (e). σg = 0.33, σt = 5, (f). σg = 1, σt = 5, (g). σg = 0.33, σt = 9, (h). σg = 1, σt = 9. 18 Figure 14. Filters in horizontal direction (a). line segment filter, (b). concave-up curve-shaped filter, (c). concave-down curve-shaped filter. 19 Figure 15. Comparison of line drawings based on different filters (a). original image, (b). C. Lu et al., (c). LDrawing 1, (d). LDrawing 2, (e). LDrawing 3, (f).~(j). close-ups for (a)~(e). 20 Figure 16. An example of LDrawing 3 (a). original image, (b). result. 21 Figure 17. Examples of generated pencil texture (a). θ = 0, (b). θ = π/4, (c). combination of (a) and (b). 22 Figure 18. Transparency adjustment for texture image given in Fig. 16(c) with (a). α = 0. 25, (b). α = 0.5, (c). α = 0.75 22 Figure 19. An example of modified shock filter (a). original image, (b) result. 23 Figure 20. An image pyramid 24 Figure 21. Image pyramid processing 25 Figure 22. Results of image pyramid processing (a). original image, (b)~(d). results of 1, 2, and 3 iterations of pyramid processing, respectively. 25 Figure 23. Image abstraction (a). result of optional color quantization for Fig. 20(d), (b). result of adding edges for Fig. 21(a). 26 Figure 24. Comparison of three versions of pencil sketch (a). original image, (b).~(d). results of LDrawing 1, LDrawing 2, and LDrawing 3, respectively, (e).~(g). pencil sketches based on LDrawing 1, LDrawing 2, and LDrawing 3, respectively, (h). a modified version of (g). 27 Figure 25. An example of effect of watercolor painting 28 Figure 26. Results of CGP (a). original image, (b). abstraction result of (a), (c).&(d). vector fields of (a)&(b), (e).&(f). CGP results of (a)&(b). 29 Figure 27. Example of modified CGP (a). CGP result given in Fig. 7(a), (b). the flow-based Gaussian blurring result of (a), (c). shock filtering result of (b). 30 Figure 28. Example of modified CGP (a). original image, (b). CGP result of (a), (c). the flow-based Gaussian blurring result of (b), (d). shock filtering result of (c). 30 Figure 29. Results of abstraction (a).&(b). original images, (c).&(d). 3-iteration pyramid processing results of (a)&(b), (e).&(f). OCQ results of (c)&(d), (g).&(h). edge adding results of (e)&(f). 31 Figure 30. Examples of pencil sketch (a).&(e). original images, (b).&(f). pencil sketch using LDrawing 1, (c).&(g). pencil sketch using LDrawing 2, (d).&(h). pencil sketch using LDrawing 3. 32 Figure 31. Example of the color pencil sketch (a). original image, (b). result. 33 Figure 32. Results of watercolor painting (a). & (c). original images, (b). & (d). results of (a) and (c). 33 Figure 33. Results of some other effects (a).&(d). original images, (b).&(f). flow-based Gaussian filtering, (c).&(g). shock filtering (Gaussian σ = 1.2, 5 iterations), (d).&(h). flow-based Gaussian filtering + shock filtering 34 Figure 34. Examples of CGP (a).&(c). original images, (b).&(d). results. 34 Figure 35. Example of modified CGP (a).&(e).&(i). original images, (b).&(f).&(j). CGP (b).&(f).&(j). CGP + the flow-based Gaussian filtering, (d).&(h).&(l). CGP + the flow-based Gaussian filtering + shock filtering. 35 Figure 36. Examples of more effects (a). &(d).&(g).&(j). original images, (b).&(e).&(h).&(k). shock filtering, (c).&(f).&(i).&(l). modified shock filtering + flow-based Gaussian filtering. 36 Figure 37. The interface of the proposed system 37 |
參考文獻 |
[1] Y. Boykov and M.-P. Jolly, “Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in n-D Images,” in Proceedings of the IEEE International Conference on Computer Vision, Vol.1, pp. 105-112, 2001. [2] X. Bai and G. Sapiro, “A Geodesic Framework for Fast Interactive Image and Video Segmentation and Matting,” in Proceedings of the IEEE International Conference on Computer Vision, pp. 1-8, 2007. [3] 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. [4] Y. Bin and A.-D. Saleh, "Structure Enhancing Bilateral Filtering of Images," in Proceedings of the 2010 First International Conference on Pervasive Computing, Signal Processing and Applications, pp. 614-617, 2010. [5] B. Gooch, E. Reinhard, and A. Gooch, “Human Facial Illustrations: Creation and Psychophysical Evaluation,” ACM Transactions on Graph, Vol. 23, No. 1, pp. 27-44, 2004. [6] G. Papari and N. Petkov, “Continuous Glass Patterns for Painterly Rendering,” IEEE Transactions on Image Processing, Vol. 18, No. 3, pp. 652-664, 2009. [7] A. Bousseau, F. Neyret, J. Thollot, and D. Salesin, “Video Watercolorization Using Bidirectional Texture Advection,” in Proceedings of ACM SIGGRAPH, Vol. 26, No. 3, Article No. 104, 2007. [8] H. Winnemoller, S. C. Olsen, and B. Gooch, “Real Time Video Abstraction,” in Proceedings of ACM SIGGRAPH, pp. 1221-1226, 2006. [9] H. Kang, S. Lee, and C. K. Chui, “Coherent Line Drawing,” in Proceedings of the 5th International Symposium on Non-photorealistic Animation and Rendering, pp. 43-50, 2007. [10] Y. T. Kao, H. J. Lin, and C. M. Chiang, “Flexible Colorization for Greyscale Videos,” Applied Mechanics and Materials, Vols. 229-231, pp. 1495-1498, 2012. [11] A. Levin, D. Lischinski, and Y. Weiss, “Colorization Using Optimization,” ACM Transactions on Graphics, pp. 689-694, 2004. [12] H. J. Lin, J. P. Yeh, and C. Y. Wu, “Optimal Colorization for Grayscale Images,” Imaging Science Journal, Vol. 57, No. 2, pp. 61-68, 2009. [13] M. Brown, R. Szeliski, and S. Winder, “Multi-Image Matching Using Multi-Scale Oriented Patches,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Vol. 1, pp. 510-517, 2005. [14] A. Agarwala, M. Dontcheva, M. Agarwala , A. Druker, A. Colburn, B. CURLESS, D. SALESIN, and M. COHEN, “Interactive Digital Photomontage,” ACM Transactions on Graphics, Vol. 23 No. 3, pp. 294-302, 2004. [15] D. Lischinski, Z. Farbman, M. Uyttendaele, and R. Szeliski, “Interactive Local Adjustment of Tonal Values,” ACM Transactions on Graphics, Vol. 25, No. 3, pp. 646-653, 2006. [16] J. B. Weiss, “Fast Median and Bilateral Filtering,” in ACM Transactions on Graphics, Vol. 25, No. 3, pp. 646-653, 2006. [17] J. E. Kyprianidis, H. Kang, and J. Dollner, “Image and Video Abstraction by Anisotropic Kuwahara Filtering,” Computer Graphics Forum, Vol. 28, No. 7, pp. 1955-1963, 2009. [18] Y. Zhao, D. E. Xie, and D. Xu, “GPU Based Video Stylization,” Machine Learning and Cybernetics, Vol. 5, pp. 2880-2885, 2008. [19] G. Papari, N. Petkov, and P. Campisi, “Artistic Edge and Corner Enhancing Smoothing,” IEEE Transactions on Image Processing, Vol. 29, No. 10, pp. 2449-2462, 2007. [20] J. E. Kyprianidis and J. Dollner, “Image Abstraction by Structure Adaptive Filtering,” in Proceedings of EG UK Theory and Practice of Computer Graphics, pp. 51-58, 2008. [21] J. E. Kyprianidis, "Image and Video Abstraction by Multi-scale Anisotropic Kuwahara Filtering," in Proceedings of 9th Symposium on Non-Photorealistic Animation and Rendering, pp. 55-64, 2011. [22] J. E. Kyprianidis and H. Kang, “Image and Video Abstraction by Coherence-Enhancing Filtering, “Computer Graphics Forum, Vol. 30, No. 2, pp. 593-602, 2011. [23] D. E. Xie, Y. Zhao, and D. Xu, “An Efficient Approach for Generating Pencil Filter and Its Implementation on GPU,” Computer-Aided Design and Computer Graphics, pp. 185-190, 2007. [24] C. Lu, L. Xu, and J. Jia, “Combining Sketch and Tone for Pencil Drawing Production,” In Proceeding of International Symposium on Non-Photorealistic Animation and Rendering, pp. 65-73, 2012. [25] H. Winnemoller, “XDoG: Advanced Image Stylization with eXtended Difference-of-Gaussians,” in Proceedings of the ACM SIGGRAPH/Eurographics Symposium on Non-Photorealistic Animation and Rendering, pp. 147-156, 2011. [26] E. Lei and C.-F. Chang, “Real-Time Rendering of Watercolor Effects for Virtual Environments,” in Proceedings of IEEE 2004 Pacific Rim Conference on Multimedia, pp. 474-481, 2004. [27] L. Glass, “Moire Effect from Random Dots,” Nature, Vol. 223, pp. 578-580, 1969. [28] S. Osher and L. I. Rudin, “Feature-Oriented Image Enhancement Using Shock Filters,” SIAM Journal on Numerical Analysis, Vol. 27, No. 4, pp.919-940, 1990. [29] F. Guichard and J.-M. Morel, “A Note on Two Classical Shock Filters and Their Asymptotics,” In M. Kerckhove, editor, Scale-Space and Morphology in Computer Vision, volume 2106 of Lecture Notes in Computer Science, pp. 75-84, Springer, Berlin, 2001. [30] Joachim Weickert, “Coherence-Enhancing Shock Filters,” In B. Michaelis, G. Krell (Eds.): Pattern Recognition. Lecture Notes in Computer Science, Vol. 2781, pp. 1-8, Springer, Berlin, 2003. [31] J. Bigun and G. H. Granlund, “Optimal Orientation Detection of Linear Symmetry,” in Proceedings First International Conference on Computer Vision, IEEE Computer Society Press, London, England, pp. 433-438, 1987. [32] T. Q. Pham and L. J. van Vliet, “Separable Bilateral Filtering for Fast Video Preprocessing,” In Proceedings of the IEEE International Conference on Multimedia and Expo, 2005. [33] D. DeCarlo and A. Santella, “Stylization and Abstraction of Photographs,” ACM Transactions on Graphics, Vol. 21, No. 3, pp. 769-776, 2002. [34] V. Lakshmanan, “A Separable Filter for Directional Smoothing,” IEEE Geoscience and Remote Sensing Letters, Vol. 1, pp. 192-195, 2004. |
論文全文使用權限 |
如有問題,歡迎洽詢!
圖書館數位資訊組 (02)2621-5656 轉 2487 或 來信