Abstract | ||
---|---|---|
Subspace-based face representation can be looked as a regression problem. From this viewpoint, we first revisited the problem of recognizing faces across pose differences, which is a bottleneck in face recognition. Then, we propose a new approach for cross-pose face recognition using a regressor with a coupled bias-variance tradeoff. We found that striking a coupled balance between bias and variance in regression for different poses could improve the regressor-based cross-pose face representation, i.e., the regressor can be more stable against a pose difference. With the basic idea, ridge regression and lasso regression are explored. Experimental results on CMU PIE, the FERET, and the Multi-PIE face databases show that the proposed bias-variance tradeoff can achieve considerable reinforcement in recognition performance. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/TIP.2011.2160957 | IEEE Transactions on Image Processing |
Keywords | Field | DocType |
subspace-based face representation,regression problem,multi-pie face databases,face recognition,proposed bias,lasso regression,regressor-based cross-pose face representation,cross-pose face recognition,ridge regression,variance tradeoff,accuracy,face,solid modeling,three dimensional,feature extraction | Bottleneck,Computer science,Lasso (statistics),Bias–variance tradeoff,Artificial intelligence,Computer vision,Facial recognition system,Regression,Pattern recognition,Subspace topology,Feature extraction,Solid modeling,Machine learning | Journal |
Volume | Issue | ISSN |
21 | 1 | 1941-0042 |
Citations | PageRank | References |
73 | 1.97 | 23 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Annan Li | 1 | 222 | 14.22 |
Shiguang Shan | 2 | 6322 | 283.75 |
Wen Gao | 3 | 11374 | 741.77 |