원문정보
초록
영어
In this paper, we propose a new method of image deformation. The unequal grids, which are based on the density of edges, were used for the dividing criteria in the MLS (Moving Least Squares) approximation [1] in image deformation. Depending on the importance of the image, we place the size of the grid differently and minimize the distortion that can occur when transforming the image. In this method, we introduce the image partition method using the quad tree for densely placing the small mesh in complex part with lots of edge in the image. By using this method, we can improve the quality of the images and increase the processing speed of image deformation. The validity of this method will be shown by the presentation of the results of deformed images with unequal grids.
목차
1. Introduction
2. Related Work
3. System Overview
3.1. Pre-compute: Edge Detection
3.2. Quad-tree Segmentation using the Density of Edges
3.3. Moving Least Squares Deformation
4. Results
5. Conclusion
Acknowledgements
References