Real World Heart Rate Estimation from Video
| Project/Area Number |
17K12709
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | Saitama University |
|---|
Principal Investigator |
Lam Antony 埼玉大学, 理工学研究科, 助教 (50744124)
|
|---|
| Project Period (FY) |
2017-04-01 – 2019-03-31
|
| Project Status |
Completed (Fiscal Year 2018)
|
| Budget Amount *help |
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2018: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | video ppg / computer vision / heart rate / cardiac estimation / remote measurement / 画像、文章、音声認識等 / 知能ロボティックス / Heart Rate / Monitoring / Computer Vision |
|---|
| Outline of Final Research Achievements |
Our aim in this project was to develop an algorithm that can take a video feed of someone's face and through reading subtle changes in skin color (imperceptible to the human eye), estimate the person's heart rate. We took our previous work that was robust to lighting changes in real-world environments and improved it by adding robustness to out-of-pose head rotations. This was achieved by tracking triangular local regions of the face where the regions could adaptively change shape to fit many 3D rotations of the head. The tracked triangular local regions of the face were then used to perform a robust estimate of cardiac activity from the subtle color changes of the skin. We found this technique works even with RGB webcams. We also developed a variant of the algorithm that is faster but less accurate so it could be used for different applications. This variant worked by considering specific face regions with strong cardiac-caused color changes in skin.
|
| Academic Significance and Societal Importance of the Research Achievements |
Our work is one of the few in the literature that addressed the difficult challenges of estimating cardiac activity from videos with changing illumination and motion in the real-world. This algorithm can also be used for applications such as emotion recognition and human computer interaction.
|
Report
(3 results)
Research Products
(7 results)
