2000 Fiscal Year Final Research Report Summary
Modeling attention shift mechanism and its applications to adaptive computer vision
| Project/Area Number |
10480074
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B).
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | OKAYAMA UNIVERSITY |
|---|
Principal Investigator |
SHAKUNAGA Takeshi Faculty of engineering, OKAYAMA UNIVERSITY, Professor, 工学部, 教授 (80284082)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MUKAIGAWA Yasuhiro Graduate school of natural science and technology, OKAYAMA UNIVERSITY, Research associate, 大学院・自然科学研究科, 助手 (60294435)
|
|---|
| Project Period (FY) |
1998 – 2000
|
| Keywords | Attention control / Image processin / Adaptive vision / Person tracking / Motion analysis / State space / Hierarchical attention control / Computational model |
|---|
| Research Abstract |
(a) Hierarchical attention control model :
A hierarchical attention control model, which consists of five layers is proposed for implementing a human-like attention control in computer vision. The lowest layer is called a window layer, in which each attention window is controlled by signal processing in the same manner as in the conventional attention control. The other four layers, however, are designed for high-level control, of which the top layer is used only for the mission control. In our model, the attention control is accomplished using the tree structure across the five layers and both the bottom-up and top-down processings are accomplished on it.
(b) Application to person tracking in natural environment
An application of the model is shown for the person tracking in complex situations. In this application, multiple windows are generated on the image plane and processed in the window layer. A priority is attached to each window, which shows the effectiveness for the tracking. The priority is controlled in the object layer.
(c) Application to stereoscopic human motion analysis
An application of the model is shown for human motion analysis. In this application, correspondence between human structure model and stereoscopic data for motion interpretation.
|
