Gaze Estimation based on Saliency Dynamics Analysis
| Project/Area Number |
15K16018
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
Yonetani Ryo 東京大学, 生産技術研究所, 助教 (40733481)
|
|---|
| Project Period (FY) |
2015-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2016: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 視線推定 / キャリブレーション / 顕著性 |
|---|
| Outline of Final Research Achievements |
We develop a technique to estimate where people look at when they are watching video contents. The proposed method makes use of a property of videos called 'saliency dynamics', which models how salient regions in videos dynamically move over time, for gaze estimation. Specifically, we propose to analyze the correlation between saliency dynamics and eye movements. Intuitively, when people pursue moving objects in a video, we can observe a positive correlation between object motion and eye movements. This correlation feature allows us gaze estimation without manual calibration procedures.
Experimental results have demonstrated the effectiveness of the proposed approach. These results were presented at two major computer vision conferences in Japan.
|
Report
(3 results)
Research Products
(2 results)
