Optimal metric selection for improved multi-pose face recognition with group information
We address the limitation of sparse representation based classification with group information for multi-pose face recognition. First, we observe that the key issue of such classification problem lies in the choice of the metric norm of the residual vectors, which represent the fitness of each class...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Published: |
ICPR
2012
|
| Subjects: | |
| Online Access: | http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460470 http://hdl.handle.net/20.500.11937/46687 |
| _version_ | 1848757629335109632 |
|---|---|
| author | Zhang, X. Pham, DucSon Liu, W. Venkatesh, S. |
| author2 | N/A |
| author_facet | N/A Zhang, X. Pham, DucSon Liu, W. Venkatesh, S. |
| author_sort | Zhang, X. |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | We address the limitation of sparse representation based classification with group information for multi-pose face recognition. First, we observe that the key issue of such classification problem lies in the choice of the metric norm of the residual vectors, which represent the fitness of each class. Then we point out that limitation of the current sparse representation classification algorithms is the wrong choice of the ℓ2 norm, which does not match with data statistics as these residual values may be considerably non-Gaussian. We propose an explicit but effective solution using ℓp norm and explain theoretically and numerically why such metric norm would be able to suppress outliers and thus can significantly improve classification performance comparable to the state-of-arts algorithms on some challenging datasets. |
| first_indexed | 2025-11-14T09:31:08Z |
| format | Conference Paper |
| id | curtin-20.500.11937-46687 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T09:31:08Z |
| publishDate | 2012 |
| publisher | ICPR |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-466872017-01-30T15:28:42Z Optimal metric selection for improved multi-pose face recognition with group information Zhang, X. Pham, DucSon Liu, W. Venkatesh, S. N/A vectors training face face recognition lighting robustness measurement We address the limitation of sparse representation based classification with group information for multi-pose face recognition. First, we observe that the key issue of such classification problem lies in the choice of the metric norm of the residual vectors, which represent the fitness of each class. Then we point out that limitation of the current sparse representation classification algorithms is the wrong choice of the ℓ2 norm, which does not match with data statistics as these residual values may be considerably non-Gaussian. We propose an explicit but effective solution using ℓp norm and explain theoretically and numerically why such metric norm would be able to suppress outliers and thus can significantly improve classification performance comparable to the state-of-arts algorithms on some challenging datasets. 2012 Conference Paper http://hdl.handle.net/20.500.11937/46687 http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460470 ICPR fulltext |
| spellingShingle | vectors training face face recognition lighting robustness measurement Zhang, X. Pham, DucSon Liu, W. Venkatesh, S. Optimal metric selection for improved multi-pose face recognition with group information |
| title | Optimal metric selection for improved multi-pose face recognition with group information |
| title_full | Optimal metric selection for improved multi-pose face recognition with group information |
| title_fullStr | Optimal metric selection for improved multi-pose face recognition with group information |
| title_full_unstemmed | Optimal metric selection for improved multi-pose face recognition with group information |
| title_short | Optimal metric selection for improved multi-pose face recognition with group information |
| title_sort | optimal metric selection for improved multi-pose face recognition with group information |
| topic | vectors training face face recognition lighting robustness measurement |
| url | http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6460470 http://hdl.handle.net/20.500.11937/46687 |