在本論文的研究中，我們為動態影像形狀雲的視覺化研究中帶入了更加動態的元素。影像形狀雲是來自於文字雲的視覺化藝術，資料視覺化的方法能夠使大眾更方便且快速的了解作者想傳達的訊息，相對於文字，影像又能帶來更豐富且有趣的效果，對於主題相關聯的影像形狀雲，彼此能夠形成動態的影像形狀雲，更生動的傳達內容。
影像形狀雲的研究上，對於不規則的影像物件我們透過記數圖以及物件輪廓圖達到不浪費空間的拼貼，並且針對外型空間進行量化，使影像的重要度不僅影響影像的大小，並且在空間中找到合乎重要度的拼貼位置。此外我們透過二維三角形網格(2D triangular mesh)在系統中代理影像物件，使網格能夠模擬質量-彈簧(mass-spring)系統呈現影像的變形。藉由動態模擬的方式使網格填充空間，相比之前影像形狀雲的研究大幅提高空間使用率，且不使用細小物件填充空間，形成更加的資料視覺化效果。動態影像形狀雲的研究中，一改以往透過剛性模擬的方法中除了影像物件轉換過程中會穿透外型、大量消失及冒出，物件也較為死板。本論文的方法使影像能在影像形狀雲轉換時自由的移動、旋轉或變形，且不超出外型邊界，產生具有動態感且不突兀的動畫過程，形成更生動的動態影像形狀雲。

In this research, we introduced more dynamic elements into the visualization of dynamic 2D(image) shape clouds. Shape clouds are a form of visual art derived from word clouds, where data visualization techniques enable the audience to conveniently and quickly understand the message conveyed by the author. Compared to text, images bring more richness and interest, and for topic-related shape clouds, they can form dynamic shape clouds that convey content more vividly.
In the study of shape clouds, for irregular image objects, we achieve efficient collage by using count maps and object contour maps, without wasting space. We also quantify the shape canvas space to ensure that the importance of images not only affects their size but also determines suitable collage positions in the space. Furthermore, we employ a 2D triangular mesh to represent image objects in the system, allowing the mesh to simulate mass-spring systems and exhibit image deformation. By filling the space with the mesh through physical simulation, the research significantly improves space utilization compared to previous shape cloud studies, without relying on small objects to fill the space, resulting in enhanced data visualization effects.
In the study of dynamic 2D(image) shape clouds, we depart from the previous rigid simulation approach, which often led to penetration of the canvas shape space, the disappearance or sudden appearance of objects, and rigid object behavior during the transformation of image objects. The method proposed in this paper enables images to freely move, rotate, or deform during the transformation of shape clouds, while staying within the canvas shape boundaries. This produces dynamic animations with a sense of motion and smooth transitions, creating more lively and dynamic 2D shape clouds.

[1] Y.-H. Tsai and T.-Y. Lee, "Image Shape Collage and Time-varying Image Shape Cloud Visualization," Dissertations and Theses, National Cheng Kung University, College of Electrical Engineering and Computer Science/ Graduate Program of Artificial Intelligence, 2018.
[2] Wu, Dong-Yi, Sheng-Yi Yao, Yun-Chen Lin, and Tong-Yee Lee. "Image Collage on Arbitrary Shape via Shape-Aware Slicing and Optimization." IEEE Transactions on Visualization and Computer Graphics (2023).
[3] Wang, Yu-Shuen, Chaoli Wang, Tong-Yee Lee, and Kwan-Liu Ma. "Feature-preserving volume data reduction and focus+ context visualization." IEEE Transactions on Visualization and Computer Graphics 17, no. 2 (2010): 171-181.
[4] C.-L. Chia and T.-Y. Lee, " Time Varying Visualization for Image Shape Cloud with Neural Network," Dissertations and Theses, National Cheng Kung University, College of Electrical Engineering & Computer Science / Graduate Program of Artificial Intelligence, 2021.
[5] Viegas, Fernanda B., Martin Wattenberg, and Jonathan Feinberg. "Participatory visualization with wordle." IEEE transactions on visualization and computer graphics 15, no. 6 (2009): 1137-1144.
[6] Koh, Kyle, Bongshin Lee, Bohyoung Kim, and Jinwook Seo. "Maniwordle: Providing flexible control over wordle." IEEE Transactions on Visualization and Computer Graphics 16, no. 6 (2010): 1190-1197.
[7] Strobelt, Hendrik, Marc Spicker, Andreas Stoffel, Daniel Keim, and Oliver Deussen. "Rolled‐out wordles: A heuristic method for overlap removal of 2d data representatives." In Computer Graphics Forum, vol. 31, no. 3pt3, pp. 1135-1144. Oxford, UK: Blackwell Publishing Ltd, 2012.
[8] Maharik, Ron, Mikhail Bessmeltsev, Alla Sheffer, Ariel Shamir, and Nathan Carr. "Digital micrography." ACM Transactions on Graphics (TOG) 30, no. 4 (2011): 1-12.
[9] Wang, Yunhai, Xiaowei Chu, Kaiyi Zhang, Chen Bao, Xiaotong Li, Jian Zhang, Chi-Wing Fu, Christophe Hurter, Oliver Deussen, and Bongshin Lee. "Shapewordle: tailoring wordles using shape-aware archimedean spirals." IEEE Transactions on Visualization and Computer Graphics 26, no. 1 (2019): 991-1000.
[10] Cui, Weiwei, Yingcai Wu, Shixia Liu, Furu Wei, Michelle X. Zhou, and Huamin Qu. "Context preserving dynamic word cloud visualization." In 2010 IEEE Pacific Visualization Symposium (PacificVis), pp. 121-128. IEEE, 2010.
[11] Seyfert, Martin, and Ivan Viola. "Dynamic word clouds." In Proceedings of the 33rd Spring Conference on Computer Graphics, pp. 1-8. 2017.
[12] Chi, Ming-Te, Shih-Syun Lin, Shiang-Yi Chen, Chao-Hung Lin, and Tong-Yee Lee. "Morphable word clouds for time-varying text data visualization." IEEE transactions on visualization and computer graphics 21, no. 12 (2015): 1415-1426.
[13] Lee, Man Hee, Nitin Singhal, Sungdae Cho, and In Kyu Park. "Mobile photo collage." In 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition-Workshops, pp. 24-30. IEEE, 2010.
[14] Wang, Jingdong, Long Quan, Jian Sun, Xiaoou Tang, and Heung-Yeung Shum. "Picture collage." In 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06), vol. 1, pp. 347-354. IEEE, 2006.
[15] Han, Xintong, Chongyang Zhang, Weiyao Lin, Mingliang Xu, Bin Sheng, and Tao Mei. "Tree-based visualization and optimization for image collection." IEEE transactions on cybernetics 46, no. 6 (2015): 1286-1300.
[16] Huang, Hua, Lei Zhang, and Hong-Chao Zhang. "Arcimboldo-like collage using internet images." In Proceedings of the 2011 SIGGRAPH Asia Conference, pp. 1-8. 2011.
[17] Peng, Chi-Han, Yong-Liang Yang, and Peter Wonka. "Computing layouts with deformable templates." ACM Transactions on Graphics (TOG) 33, no. 4 (2014): 1-11.
[18] Kuo, Ying-Miao, Hung-Kuo Chu, Ming-Te Chi, Ruen-Rone Lee, and Tong-Yee Lee. "Generating ambiguous figure-ground images." IEEE transactions on visualization and computer graphics 23, no. 5 (2016): 1534-1545.
[19] Kwan, Kin Chung, Lok Tsun Sinn, Chu Han, Tien-Tsin Wong, and Chi-Wing Fu. "Pyramid of arclength descriptor for generating collage of shapes." ACM Trans. Graph. 35, no. 6 (2016): 229-1.
[20] Saputra, Reza Adhitya, Craig S. Kaplan, and Paul Asente. "Improved deformation-driven element packing with repulsionPak." IEEE transactions on visualization and computer graphics 27, no. 4 (2019): 2396-2408.
[21] Wang, Yunhai, Xiaowei Chu, Chen Bao, Lifeng Zhu, Oliver Deussen, Baoquan Chen, and Michael Sedlmair. "Edwordle: Consistency-preserving word cloud editing." IEEE transactions on visualization and computer graphics 24, no. 1 (2017): 647-656.
[22] Ogniewicz, Robert L., and Markus Ilg. "Voronoi skeletons: theory and applications." In CVPR, vol. 92, pp. 63-69. 1992.
[23] Müller, Matthias, Nuttapong Chentanez, Tae-Yong Kim, and Miles Macklin. "Air meshes for robust collision handling." ACM Transactions on Graphics (TOG) 34, no. 4 (2015): 1-9.
[24] Pfaff, Tobias, Meire Fortunato, Alvaro Sanchez-Gonzalez, and Peter W. Battaglia. "Learning mesh-based simulation with graph networks." arXiv preprint arXiv:2010.03409 (2020).
[25] Sanchez-Gonzalez, Alvaro, Jonathan Godwin, Tobias Pfaff, Rex Ying, Jure Leskovec, and Peter Battaglia. "Learning to simulate complex physics with graph networks." In International conference on machine learning, pp. 8459-8468. PMLR, 2020.
[26] Sarukkai, Vishnu, Linden Li, Arden Ma, Christopher Ré, and Kayvon Fatahalian. "Collage diffusion." arXiv preprint arXiv:2303.00262 (2023).
[27] Liu, Luping, Zijian Zhang, Yi Ren, Rongjie Huang, Xiang Yin, and Zhou Zhao. "Detector Guidance for Multi-Object Text-to-Image Generation." arXiv preprint arXiv:2306.02236 (2023).