DocumentCode
2546586
Title
Detection of critical points for shape metamorphosis animation
Author
Nieda, Tomoyuki ; Pasko, Alexander ; Kunii, Tosiyasu L.
Author_Institution
Dept. of digital media, Hosei Univ., Tokyo, Japan
fYear
2004
fDate
5-7 Jan. 2004
Firstpage
93
Lastpage
100
Abstract
We apply topological analysis to functionally based shape metamorphosis. The time-dependent shape is defined using homotopy. The advantage of this method is the automatic generation of the intermediate shapes between the key shapes of different topology types. To complete the method, we have to find a way to automatically detect the critical points on the time axis while the shape undergoes topological changes. These critical points can be later used for generation of non-linear time steps distribution along the time axis, for example, for providing ease-in/ease-out effects in animation. We present a new method for analysis of shape metamorphosis based on the Morse theory, oriented to analysis of a height function. Although we analyze the shape in an N-dimensional space, the height function is defined in the N+2 dimensional space with N point coordinates and two additional coordinates of the defining function and time values. We can analyze how the critical points are changing in the given height level, which takes only zero value of the shape defining function. In this paper, we present this method in comparison with typical Morse theory analysis using simple objects in 2D and 3D spaces.
Keywords
computational geometry; computer animation; image morphing; Morse theory; critical points detection; ease-in-ease-out effects; functionally based shape metamorphosis; homotopy functions; shape metamorphosis animation; topological analysis; Animation; Face detection; Information analysis; Information science; Motion pictures; Object detection; Shape; Solids; TV; Topology;
fLanguage
English
Publisher
ieee
Conference_Titel
Multimedia Modelling Conference, 2004. Proceedings. 10th International
Print_ISBN
0-7695-2084-7
Type
conf
DOI
10.1109/MULMM.2004.1264972
Filename
1264972
Link To Document