Shape correspondence based on Kendall shape space and RAG for 2D animation
- PDF / 3,057,848 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 11 Downloads / 253 Views
ORIGINAL ARTICLE
Shape correspondence based on Kendall shape space and RAG for 2D animation Shaolong Liu1
· Xingce Wang1 · Zhongke Wu1 · Hock Soon Seah2
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract We introduce a 2D vectorized shape correspondence method based on Kendall shape space and region adjacency graph. We regard shape correspondence in 2D animation as a weighted bipartite graph matching problem and optimize it by means of various weights, including geometric attribute, local topological information, and global topological information in 2D animation drawing. The measuring method in Kendall shape space is introduced to determine similarity between two regions and a local adjacent region matching method is presented to associate the vicinity of the corresponding region by utilizing local topology information. Ultimately, we propose a globally optimal shape matching method, which exploits the global topological information to get the final result of the shape correspondence. Our approach can efficiently match associated regions that have exaggerated deformation and unstable topology between two shapes in 2D animation. It is also robust to the similarity transformation of shapes. Keywords Vectorized 2D animation · Shape correspondence · Kendall shape space · Region adjacency graph
1 Introduction Up to now, 2D hand-drawn animation is still an essential part of the global entertainment industry. However, even now with computer support, some production processes such as inbetweening, coloring, objects identification, etc. are still time consuming and labor intensive [14]. Shape correspondence is a fundamental component of the aforementioned processes. The primary cause for high costs of these processes is the correspondence problem between two different shapes, which has not been solved very well even with contemporary computer-assisted technology. In 2D animation, a keyframe contains a shape, and a shape consists of multiple regions. The difficulties of shape correspondence are that the regions usually deform heavily between different keyframes by unconstrained creative drawing and physically based deformation [2] or that the topology may be inconsistent due to partial occlusion as Fig. 1 shows. Either of the above situations will seriously reduce the
B
Zhongke Wu [email protected]
1
Beijing Normal University, Beijing, China
2
School of Computer Science and Engineering, Nanyang Technological University (NTU), Singapore, Singapore
accuracy of shape correspondence that even sophisticated methods are ineffective. Therefore, shape correspondence in 2D animation is a challenging task. To solve the above-mentioned issues, previous 2D shape correspondence methods focused on how to mine and utilize geometric and topological features, respectively. Some methods utilized mainly geometric features of the regions in the shape such as area [14], position [10], various shape descriptors [5], etc. These methods rely exclusively on the appearance feature such as the area, position, curvature cha
Data Loading...
