Title | ||
---|---|---|
Geometry-Based Facial Expression Recognition Via Large Deformation Diffeomorphic Metric Curve Mapping |
Abstract | ||
---|---|---|
We proposed a new geometry-based facial expression recognition (FER) system in the framework of large deformation diffeomorphic metric curve mapping. The geometry of a face was represented by 12 distinct curves, with curve-based facial deformations being used to identify two sets of geometric features in two settings. In each setting, four types of features were extracted and tested. Leave-one-out cross-validation experiments on 327 image sequences yielded accuracies as high as 94%. Furthermore, using a multi-kernel technique to combine features from the two settings has boosted the recognition accuracy to be as high as 95.4%. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/ICIP.2018.8451764 | 2018 25TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP) |
Keywords | Field | DocType |
Facial Expression Recognition, Shape, Large Deformation Diffeomorphic Mapping, Curve | Computer vision,Facial expression recognition,Pattern recognition,Computer science,Feature extraction,Artificial intelligence,Deformation (mechanics),Geometry,Trajectory,Diffeomorphism | Conference |
ISSN | Citations | PageRank |
1522-4880 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pucheng Yang | 1 | 0 | 0.34 |
Huilin Yang | 2 | 0 | 0.34 |
Yuanyuan Wei | 3 | 0 | 0.68 |
Xiaoying Tang | 4 | 8 | 8.79 |