Abstract | ||
---|---|---|
An objective evaluation method to analyze the degree of motion dysfunction in facial expressions due to paralysis is presented. The analysis are based on range data captured for pre-determined facial actions of patients with facial paralysis. The dysfunctions in expressions are analyzed by estimating the degree of asymmetry between left and right sides of 3D expression models that were constructed from measured range data. A symmetric generic face mesh is adapted to the range data using a least squares approximation method to generate these 3D models. The 3D expression models are interpreted as inter-connected meshes of linear springs with point masses and their deformations during expressions are determined by estimating the change of energy in each case. The results obtained for patients in consecutive measurements are compared against that of normal subjects to illustrate the degree of recovery quantitatively. |
Year | DOI | Venue |
---|---|---|
2003 | 10.1007/978-3-540-39903-2_1 | Lecture Notes in Computer Science |
Keywords | Field | DocType |
facial expression,least squares approximation | Least squares,Polygon mesh,Pattern recognition,Expression (mathematics),Facial expression,Artificial intelligence,Facial paralysis,Asymmetry,Paralysis,Mathematics | Conference |
Volume | ISSN | Citations |
2879 | 0302-9743 | 1 |
PageRank | References | Authors |
0.39 | 2 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pujitha Gunaratne | 1 | 17 | 2.99 |
Yukio Sato | 2 | 2 | 0.74 |