Second-generation Video Mining : Extracting Patterns for Exhaustive Retrieval of Events in Video Archive
| Project/Area Number |
20700088
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Kobe University |
|---|
Principal Investigator |
SHIRAHAMA Kimiaki Kobe University, 大学院・経済学研究科, 助教 (30467675)
|
|---|
| Project Period (FY) |
2008 – 2009
|
| Project Status |
Completed (Fiscal Year 2009)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2009: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2008: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | 映像検索 / ビデオマイニング / ラフ集合理論 / バースト検出 / ビデオオントロジー / TRECVID / 映像アーカイブ / 多重対応分析 / バースト |
|---|
| Research Abstract |
In this research, by applying data mining technique to video processing, we have explored three approaches for efficiently retrieving events of interest in a video archive. In the first one, we use rough set theory to extract combinations of features (e.g. color, edge, motion etc.) specific to events as patterns. In the second approach, we organize various concepts (e.g. Person, Car, Building etc.) into a video ontology. It is used to improve the event retrieval performance. In the final approach, we extract impressive topics in a video by detecting abnormal editing patterns.
|
Report
(3 results)
Research Products
(23 results)
