| Project/Area Number |
14550380
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報通信工学
|
|---|
| Research Institution | Toho University |
|---|
Principal Investigator |
SHINYA Mikio Toho Univ., Faculty of Science, Professor, 理学部, 教授 (50339199)
|
|---|
| Project Period (FY) |
2002 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥2,800,000 (Direct Cost: ¥2,800,000)
Fiscal Year 2005: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 2004: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 2003: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 2002: ¥1,000,000 (Direct Cost: ¥1,000,000)
|
|---|
| Keywords | Compute Graphics / Compute Animation / Data Compression / 3D Video |
|---|
| Research Abstract |
Recent progress in digitizing technologies is making it possible to capture the 3D shapes of moving objects. Since the resulting animation data is generally huge in amount, efficient data compression is necessary to utilize such animation data. In this project, we studied compression of 3D animation data and obtained the following results.
1)Spatio-temporal compression :
We applied the progressive mesh method to reduce the number of vertices and the principal component analyses (PCA) to represent temporal changes in a compact form. Experiments showed that about 10 times data reduction can be made by each methods.
2)Unification of dynamic point data :
To apply the above method, the data should be unified into a single deforming mesh model. Unfortunately, 3D video systems generally capture independent polygonal models from frame to frame, and thus, it is necessary to unify them. We proposed an energy-based method that unifies models though energy minimization. Experiments showed the methods can deal with real data captured by a 3D video system.
3) Stabilizing energy minimization
Energy minimization can be unstable especially when the mass-spring type energy is adopted. We analyzed the simulation methods and formulated the stability conditions to ensure efficient and robust minimization.
|
