Abstract | ||
---|---|---|
In this paper, our contributions to the subspace learning problem are two-fold. We first justify that most popular subspace learning algorithms, unsupervised or supervised, can be unitedly explained as instances of a ubiquitously supervised prototype. They all essentially minimize the intraclass compactness and at the same time maximize the interclass separability, yet with specialized labeling approaches, such as ground truth, self-labeling, neighborhood propagation, and local subspace approximation. Then, enlightened by this ubiquitously supervised philosophy, we present two categories of novel algorithms for subspace learning, namely, misalignment-robust and semi-supervised subspace learning. The first category is tailored to computer vision applications for improving algorithmic robustness to image misalignments, including image translation, rotation and scaling. The second category naturally integrates the label information from both ground truth and other approaches for unsupervised algorithms. Extensive face recognition experiments on the CMU PIE and FRGC ver1.0 databases demonstrate that the misalignment-robust version algorithms consistently bring encouraging accuracy improvements over the counterparts without considering image misalignments, and also show the advantages of semi-supervised subspace learning over only supervised or unsupervised scheme. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/TIP.2008.2009415 | IEEE Transactions on Image Processing |
Keywords | Field | DocType |
ubiquitously supervised subspace learning,semi-supervised subspace,ubiquitously supervised prototype,popular subspace,unsupervised subspace learning,face recognition,image translation,local subspace approximation,ground truth,neighborhood propagation,learning (artificial intelligence),image misalignments,unsupervised subspace learning.,index terms—dimensionality reduction,dimensionality reduction,interclass separability,semi-supervised subspace learning,ubiquitously supervised philosophy,subspace learning,self-labeling,supervised subspace learning,image misalignment,intraclass compactness,linear discriminant analysis,labeling,application software,learning artificial intelligence,principal component analysis,indexing terms,computer vision,prototypes,robustness | Image translation,Dimensionality reduction,Computer science,Unsupervised learning,Artificial intelligence,Computer vision,Facial recognition system,Pattern recognition,Subspace topology,Supervised learning,Ground truth,Linear discriminant analysis,Machine learning | Journal |
Volume | Issue | ISSN |
18 | 2 | 1057-7149 |
Citations | PageRank | References |
10 | 0.50 | 17 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
jianchao yang | 1 | 7508 | 282.48 |
Shuicheng Yan | 2 | 767 | 25.71 |
Thomas S. Huang | 3 | 27815 | 2618.42 |