多視角美工圖案之輔助設計系統

Multi-view Clipart Design

劉冠宏(Kuan-Hung Liu)

指導教授 ： 陳炳宇

國立臺灣大學/管理學院/資訊管理學研究所/碩士(2020年)

https://doi.org/10.6342/NTU202002046

摘要

我們提出了一個可以輔助使用者創作不同視角美工圖案的系統。受到設計師創作過程的啟發，我們的系統會依照使用者輸入的美工圖案生成一模型，並利用該模型去渲染出使用者需要的視圖，以作為創作之視覺參照。
我們最主要的挑戰是如何讓生成的參照圖能夠符合使用者的期待，既要能夠讓生成的模型有正確的部件比例及位置，也要保有與輸入之美工圖案相似的幾何風格。因此本篇論文提出了一個使用者輔助的曲線擠出方法，並透過一致的渲染方式來生成參照。
使用者可以根據參照更有效率的在想要的視角進行創作。我們透過直觀的介面搭配生成的參照圖進行使用者研究，結果顯示透過我們的系統，使用者設計的不同視角的美工圖案在幾何風格與形狀上都與輸入之美工圖案相似。

關鍵字

參數化曲線 ； 曲面模型 ； 影像處理

並列摘要

We present an assistive system for clipart design by providing visual scaffolds fromunseen viewpoints. Inspired by the artists’ creation process, our system constructs thevisual scaffold by first synthesizing the reference 3D shape of the input clipart and ren-dering it from the desired viewpoint. The critical challengeof constructing this visual scaffold is to generate a reference 3D shape that matches theuser’s expectation in terms of object sizing and positioning while preserving the geomet-ric style of the input clipart. To address this challenge, we propose a user-assisted curveextrusion method to obtain the reference 3D shape. We render the synthesized reference3D shape with consistent style into the visual scaffold. By following the generated visualscaffold, the users can efficiently design clipart with their desired viewpoints. The userstudy conducted by an intuitive user interface and our generated visual scaffold suggeststhat the users are able to design clipart from different viewpoints while preserving theoriginal geometric style without losing its original shape.

並列關鍵字

Parametric curve ； Surface model ； Image Processing

參考文獻

Adobe. Adobe illustrator 2020: Image trace, 2020.

E. A. Ascher. Mental rotation in artists and non-artists.

S. Belongie, J. Malik, and J. Puzicha. Shape context: A new descriptor for shape matching and object recognition. In Advances in neural information processing systems, pages 831– 837, 2001.

A. X. Chang, T. Funkhouser, L. Guibas, P. Hanrahan, Q. Huang, Z. Li, S. Savarese, M. Savva, S. Song, H. Su, et al. Shapenet: An information-rich 3d model repository. arXiv preprint arXiv:1512.03012, 2015.

H.-T. Chen, T. Grossman, L.-Y. Wei, R. M. Schmidt, B. Hartmann, G. Fitzmaurice, and M. Agrawala. History assisted view authoring for 3d models. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI ’14, pages 2027–2036, New York, NY, USA, 2014. ACM.