2015 Fiscal Year Final Research Report
High-accuracy gait recognition by fluctuation analysis
| Project/Area Number |
23680017
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (A)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2011-04-01 – 2016-03-31
|
| Keywords | 歩容認証 / ゆらぎ / 位相推定 / バイオメトリクス |
|---|
| Outline of Final Research Achievements |
Since gait fluctuation often degrades the accuracy of gait recognition, we tackled with P1) gait recognition robust against the gait fluctuation. We proposed a phase estimation method from a single quasi-periodic signal as a basic techniques for this purpose, and confirmed its effectiveness with simulation and real data. Moreover, we proposed a cylindrical manifold to represent speed-transition gait and confirmed that the cylindrical manifold reconstruction successfully improve the gait recognition accuracy.
In addition, we also tackled with P2) improved gait recognition in corporation with gait fluctuation features. More specifically, we introduced phase fluctuation features and gait fluctuation image and combined them with a conventional gait feature in a score-level fusion framework and confirmed the accuracy improvement.
|
| Free Research Field |
コンピュータビジョン
|
