2016 Fiscal Year Final Research Report
Tracking and recognition of people in groups based on observations from multiple sensors
| Project/Area Number |
26330186
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Saitama University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
KUNO Yoshinori 埼玉大学, 理工学研究科, 教授 (10252595)
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Keywords | 人物行動計測 |
|---|
| Outline of Final Research Achievements |
We have established the multiple people tracking technique which enables real-time tracking of body positions and orientations of multiple users by using LiDAR sensors. Behaviors of the people in the same group are learned in advance. Then, the group of people can be recognized by the system based on the trajectories of tracking people. We conducted the experiment for evaluating the effectiveness of our method in an actual art museum. In the result distributions of residence time for each area in front of the paintings can be visualized and people trajectory patterns can be classified into four types. By using the sensors in a smartphone, our LiDAR tracking system can identify the person who is holding his/her smartphone.
|
| Free Research Field |
コンピュータビジョン
|
