組合拼貼藝術一直是許多藝術家喜愛創作的主題，其用小模塊拼貼成較大的意象往往能造成很大的視覺震撼，但要在畫面中同時保持區域和整體的可讀性是一件很不簡單的工作；另外，在視角相依視覺錯覺的影響下，要產生僅能在單一視角看的到、其他視角卻看似雜亂無章的組合拼貼則更具挑戰性。
在這篇論文研究中，我們對於快速產生視角相依三維組合拼貼提出了一個兩步驟最佳化的方法，能夠在單一視角最佳化拼貼可讀性的同時保持其他視角的模型亂度。首先藉由修改既有的模型分割演算法及三維局部匹配，產生粗略的三維模型匹配候選人，再依最佳化目標函式的三個主要限制條件解出每組匹配的評價總分並算出匹配後精細的位置。
我們的方法可以快速的實作出藝術家三模模型拼貼的效果，並可保持視角相依視覺錯覺的特性，將以往需藝術家曠日廢時的步驟快速實現。

The 3D composite art has long been welcomed by artists. With the sophisticated patch assembly it can bring great visual impact. The issue is how to keep the readability of the local patches and the global collage simultaneously. And then, how to maintain a 3D composite which can only be identified in just one view, and let other views look like merely scattering parts.
In this project, we introduce a two steps optimization approach to generate the view-dependent 3D composite. The readability of the specific viewpoint and the disorder of the remaining views should be obtained. First we use the model decomposition algorithm and the shape matching technique to produce rough matching candidates. And the second step we verify our results by performing our essential optimization process. Finally we get the optimized positions for each patch.
Our method can generate amazing 3D composite effects which keep the characteristic of the view dependent optical illusion. And it works fast.

[1] Alliez, P., Cohen-Steiner, D., Devillers, O., Lévy, B. and Desbrun, M. Anisotropic polygonal remeshing. ACM Trans. Graph., ACM, 2003, Vol. 22(3), pp. 485-493

[2] Attene, M., Falcidieno, B. and Spagnuolo, M. Hierarchical mesh segmentation based on fitting primitives. Vis. Comput., Springer-Verlag New York, Inc., 2006, Vol. 22(3), pp. 181-193

[3] Belongie, S., Malik, J. and Puzicha, J. Shape matching and object recognition using shape contexts. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 2002, Vol. 24(4), pp. 509 -522

[4] Chen, D.-Y., Tian, X.-P., Shen, Y.-T. and Ouhyoung, M. On Visual Similarity Based 3D Model Retrieval. Computer Graphics Forum, 2003, Vol. 22(3), pp. 223-232

[5] Chi, M.-T., Lee, T.-Y., Qu, Y. and Wong, T.-T. Self-Animating Images: Illusory Motion Using Repeated Asymmetric Patterns. ACM Transactions on Graphics, 2008, Vol. 27(3)

[6] Chu, H.-K., Hsu, W.-H., Mitra, N. J., Cohen-Or, D., Wong, T.-T. and Lee, T.-Y. Camouflage Images. ACM Transactions on Graphics, 2010, Vol. 29(3)

[7] Gal, R., Sorkine, O., Popa, T., Sheffer, A. and Cohen-Or, D. 3D Collage: Expressive Non-Realistic Modeling. NPAR: Proceedings of the 5th International Symposium on Non-Photorealistic Animation and Rendering ACM Press, 2007, pp. 7-14

[8] Garland, M., Willmott, A. and Heckbert, P. S. Hierarchical face clustering on polygonal surfaces. Proceedings of the 2001 symposium on Interactive 3D graphics ACM, 2001, pp. 49-58

[9] Hebert, M. and Faugeras, O. The Representation, Recognition, and Positioning of 3-D Shapes from Range Data. 3DMV87 1987, pp. 301-353

[10] Huang, H., Zhang, L. and Zhang, H.-C. Arcimboldo-like collage using internet images. ACM Trans. Graph., 2011, Vol. 30

[11] Kalvin, A. and Taylor, R. Superfaces: polygonal mesh simplification with bounded error. Computer Graphics and Applications, IEEE, 1996, Vol. 16(3), pp. 64 -77

[12] Meyer, M., Desbrun, M., Schröder, P. and Barr, A. H. Discrete Differential-Geometry Operators for Triangulated 2-Manifolds. Proc. VisMath 2002, pp. 35-57

[13] Mitra, N. J., Chu, H.-K., Lee, T.-Y., Wolf, L., Yeshurun, H. and Cohen-Or, D. Emerging Images. ACM Transactions on Graphics, 2009, Vol. 27(3)

[14] Mitra, N. J. and Pauly, M. Shadow Art. ACM Transactions on Graphics, 2009, Vol. 28(5)

[15] Pratt, V. Direct least-squares fitting of algebraic surfaces. SIGGRAPH Comput. Graph., ACM, 1987, Vol. 21(4), pp. 145-152

[16] Seckel, A. SuperVisions: Ambiguous Optical Illusions. Sterling, 2005

[17] Seckel, A. SuperVisions: Impossible Optical Illusions. Sterling, 2005

[18] Wenner, K., Hansen, B. and Hospodar, M. Asphalt Renaissance: The Pavement Art and 3-D Illusions of Kurt Wenner. Sterling Signature, 2011