Real Time 3D Facial Movement Tracking Using a Monocular Camera
Yanchao Dong
1
School of Electronics and Information Engineering, Tongji University, Caoan Road 4800, Shanghai 201804, China
2
Department of Electrical and Electronics, Kumamoto University, 2-39-1 Kurokami, Kumamoto shi 8608555, Japan
*
Author to whom correspondence should be addressed.
Academic Editor: Vittorio M. N. Passaro
Sensors 2016, 16(8), 1157; https://doi.org/10.3390/s16081157
Received: 9 May 2016 / Revised: 6 July 2016 / Accepted: 20 July 2016 / Published: 25 July 2016
(This article belongs to the Section Physical Sensors)
The paper proposes a robust framework for 3D facial movement tracking in real time using a monocular camera. It is designed to estimate the 3D face pose and local facial animation such as eyelid movement and mouth movement. The framework firstly utilizes the Discriminative Shape Regression method to locate the facial feature points on the 2D image and fuses the 2D data with a 3D face model using Extended Kalman Filter to yield 3D facial movement information. An alternating optimizing strategy is adopted to fit to different persons automatically. Experiments show that the proposed framework could track the 3D facial movement across various poses and illumination conditions. Given the real face scale the framework could track the eyelid with an error of 1 mm and mouth with an error of 2 mm. The tracking result is reliable for expression analysis or mental state inference.
View Full-Text
Keywords:
facial animation; facial feature points; 3D facial movement; eyelid; HCI
▼
Show Figures
Figure 1
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited
- Supplementary File 1:
Supplementary (ZIP, 51513 KB)
MDPI and ACS Style
Dong, Y.; Wang, Y.; Yue, J.; Hu, Z. Real Time 3D Facial Movement Tracking Using a Monocular Camera. Sensors 2016, 16, 1157.
Show more citation formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.
Search more from Scilit
