An Empirical Comparison of SNoW and SVMs for Face Detection
Impressive claims have been made for the performance of the SNoW algorithm on face detection tasks by Yang et. al. [7]. In particular, by looking at both their results and those of Heisele et. al. [3], one could infer that the SNoW system performed substantially better than an SVM-based system, even...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Online Access: | http://hdl.handle.net/1721.1/7219 |
| _version_ | 1811069907505250304 |
|---|---|
| author | Alvira, Mariano Rifkin, Ryan |
| author_facet | Alvira, Mariano Rifkin, Ryan |
| author_sort | Alvira, Mariano |
| collection | MIT |
| description | Impressive claims have been made for the performance of the SNoW algorithm on face detection tasks by Yang et. al. [7]. In particular, by looking at both their results and those of Heisele et. al. [3], one could infer that the SNoW system performed substantially better than an SVM-based system, even when the SVM used a polynomial kernel and the SNoW system used a particularly simplistic 'primitive' linear representation. We evaluated the two approaches in a controlled experiment, looking directly at performance on a simple, fixed-sized test set, isolating out 'infrastructure' issues related to detecting faces at various scales in large images. We found that SNoW performed about as well as linear SVMs, and substantially worse than polynomial SVMs. |
| first_indexed | 2024-09-23T08:18:52Z |
| id | mit-1721.1/7219 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T08:18:52Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/72192019-04-09T17:53:21Z An Empirical Comparison of SNoW and SVMs for Face Detection Alvira, Mariano Rifkin, Ryan Impressive claims have been made for the performance of the SNoW algorithm on face detection tasks by Yang et. al. [7]. In particular, by looking at both their results and those of Heisele et. al. [3], one could infer that the SNoW system performed substantially better than an SVM-based system, even when the SVM used a polynomial kernel and the SNoW system used a particularly simplistic 'primitive' linear representation. We evaluated the two approaches in a controlled experiment, looking directly at performance on a simple, fixed-sized test set, isolating out 'infrastructure' issues related to detecting faces at various scales in large images. We found that SNoW performed about as well as linear SVMs, and substantially worse than polynomial SVMs. 2004-10-20T20:50:07Z 2004-10-20T20:50:07Z 2001-01-01 AIM-2001-004 CBCL-193 http://hdl.handle.net/1721.1/7219 en_US AIM-2001-004 CBCL-193 1232391 bytes 319169 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | Alvira, Mariano Rifkin, Ryan An Empirical Comparison of SNoW and SVMs for Face Detection |
| title | An Empirical Comparison of SNoW and SVMs for Face Detection |
| title_full | An Empirical Comparison of SNoW and SVMs for Face Detection |
| title_fullStr | An Empirical Comparison of SNoW and SVMs for Face Detection |
| title_full_unstemmed | An Empirical Comparison of SNoW and SVMs for Face Detection |
| title_short | An Empirical Comparison of SNoW and SVMs for Face Detection |
| title_sort | empirical comparison of snow and svms for face detection |
| url | http://hdl.handle.net/1721.1/7219 |
| work_keys_str_mv | AT alviramariano anempiricalcomparisonofsnowandsvmsforfacedetection AT rifkinryan anempiricalcomparisonofsnowandsvmsforfacedetection AT alviramariano empiricalcomparisonofsnowandsvmsforfacedetection AT rifkinryan empiricalcomparisonofsnowandsvmsforfacedetection |