Anamorphosis中文稱為變形圖像，是一門自文藝復興早期開始出現的影像藝術。其表現技法上可分為兩種：第一類是利用反射鏡像(mirror)來表現；第二類是從視覺上的不同觀點(perspective)著手。具體來說，Anamorphosis是一種將圖像經過特殊的變形、扭曲後，使其原本表現之圖像會被扭曲投影至三維空間中之另一平面，觀看者需經由某種特定光學反射裝置(第一類)或是處於特定之觀看位置(第二類)，才能在視覺上重建回原本的圖像。
本篇論文主要是針對上述第一類的方法，建立一個系統提供使用者去創造出屬於自己的現代風格鏡射變形圖像。將想要呈現在圓柱反射鏡之圖像經過扭曲變形得到背景圖，然後自使用者限定之資料庫中全自動地將具有主題關聯性的物體填滿背景圖。根據物體的灰階值進行區塊化，透過物體本身區塊的相互關係，以及物體與背景的相互關係進行灰階值最佳化。最後透過光跡追蹤法即可在視覺上還原出原本設定之圖像，同時也保持了背景圖上物體的可辨識度。

Anamorphosis is a distorted projection requiring the viewer to use special devices or occupy a specific vantage point to reconstitute the image. The word “Anamorphosis” is derived from Greek meaning “formed again”. There are two main types of Anamorphosis: Perspective and Mirror. With mirror Anamorphosis, a cylindrical mirror is placed on the drawing or painting to transform a flat distorted image into a three dimensional picture that can be viewed from many angles. The deformed image is painted on a plane surface surrounding the mirror. By looking uniquely into the mirror, the image appears undeformed. This process of Anamorphosis made it possible to diffuse caricatures scenes.
In this paper, we construct a system that creates the mirror Anamorphosis which can’t be done with simple alpha blending. We use distorted projection to get the anamorphic input image on the background. Then, system will search figures from the user-defined database to fill up whole background automatically. According to quantization of background and figure, we manage to recolor the quantized color of figure through luminance optimization which is based on relationship between background and figure. Finally, we could see undeformed input image using the ray tracing. We believe the proposed system can provide a user-friendly tool to create mirror Anamorphosis.

[1] Chu, H.-K., Hsu, W.-H., Mitra, N.J., Cohen-Or, D., Wong, T.-T., and Lee, T.-T. (2010), “Camouflage Images”, ACM Transactions on Graphics, 29, Issue 4, pp.51:3~51:5
[2] Donoser, M., Riemenschneider, H. and Bischof, H. (2009) “Efficient partial shape matching of outer contours”, Proceedings of Asian Conference on Computer Vision, vol. 5994, pp. 281-292
[3] Gal, R., Sorkine, O., Popa, T., Sheffer, A. and Cohen-Or, D. (2007) “3D collage: expressive non-realistic modeling” Proceedings of the 5th international symposium on Non-photorealistic animation and rendering, pp.6~14
[4] Hunt, J.L., Nickel, B.G. and Gigault, C. (2000), “Anamorphic Images”, American Journal of Physics, 68, pp.232~237
[5] Kim, J., Pellacini, F., (2002), "Jigsaw Image Mosaics", ACM Transactions on Graphics, 21, pp.657~664
[6] Livinqstone, M. (2002), Vision and Art: The Biology of Seeing(1st ed.), Harry N. Abrams, pp.184~187
[7] Oliva, A., Torralba, A., Schyns, P.G. (2006), "Hybrid Images", ACM Transactions on Graphics, 25, pp.527~529
[8] Reinhard, E., Adhikhmin, M., Gooch, B. and Shirley, P. (2001) “Color Transfer between Images” IEEE Computer Graphics and Applications, 21, Issue 5, pp.35~41
[9] Rother, C., Bordeaux, L., Hamadi, Y., Blake, A., (2006) “AutoCollage” ACM Transactions on Graphics, 25, pp.847~852
[10] Smith, K., Liu Y., AND Klein, A. (2005), “Animosaics” In Proceedings of the 2005 ACM SIGGRAPH/Eurographics symposium on Computer animation, pp.201~208
[11] Wang, Z., Bovik, A.C., Sheikh, H.R. and Simoncelli, E.P. (2004), “Image Quality Assessment : From Error Visibility to Structural Similarity” IEEE Transaction on Image Processing, 13, 4, pp.600~612
[12] Yoon, J.-C., Lee, I.-K. and Kang, H. (2008) “A Hidden-picture Puzzles Generator” Pacific Graphics, 27, Issue 7, pp.1869~1877