Face recognition using modified singular-value-perturbed principal component analysis
For the past 20 years, both the academic and industrial communities have paid more attention and interest on the research and development of face recognition technologies. Generally, face recognition can be classified into three categories, i.e., holistic, feature-based and hybrid method. The most...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2010
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | For the past 20 years, both the academic and industrial communities have paid more attention and interest on the research and development of face recognition technologies. Generally, face recognition can be classified into three categories, i.e., holistic, feature-based and hybrid method. The most widely used algorithms for the holistic method are eigenface, Fisherfaces, and ICA. Face recognition techniques has numerous potential in various applications, particularly in information security, law enforcement and surveillance. Yet, a lot of these real-world applications are facing the difficulties in collecting large face samples. Nevertheless, the performance of various holistic methods is strongly affected by the number of training sample for each face and most of them will experience serious performance drop or even fail to work if only one training sample per person is available in the system. Moreover, fewer samples per person means less effort in collecting face samples, and reducing the storage and computational cost. |
|---|