Abstract | ||
---|---|---|
Pose variance remains a challenging problem for face recognition. In this paper, a stereoscopic synthesis method for generating a frontal face image is proposed to improve the performance of automatic face recognition system. Through this method, a frontal face image is generated based on two posevariant face images. Before the synthesis, face pose estimation, feature point extraction and alignment are executed on the two non-frontal images. Benefited from the high accuracy of pose estimation and alignment, the composed frontal face retains the most important features of the two corresponding nonfrontal face images. Experiment results show that using the synthetic frontal image achieves a better recognition rate than using the non-frontal ones. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1007/978-3-540-74549-5_99 | ICB |
Keywords | Field | DocType |
multiple pose-variant image,automatic face recognition system,face recognition,non-frontal image,synthetic frontal image,frontal face image,posevariant face image,frontal face synthesis,stereoscopic synthesis method,corresponding nonfrontal face image,frontal face,better recognition rate,stereoscopy,pose estimation,texture synthesis | Facial recognition system,Computer vision,Object-class detection,Pattern recognition,Three-dimensional face recognition,Face synthesis,Computer science,Stereoscopy,Pose,Artificial intelligence,Face detection,Texture synthesis | Conference |
Volume | ISSN | ISBN |
4642 | 0302-9743 | 3-540-74548-3 |
Citations | PageRank | References |
2 | 0.41 | 8 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Congcong Li | 1 | 240 | 16.48 |
Guangda Su | 2 | 133 | 20.68 |
Yan Shang | 3 | 48 | 4.04 |
Yingchun Li | 4 | 5 | 1.50 |