• Home

• Definition
• Orientation selectivity
• Computation
• Applications

• Bibliography

• Downloads

• Contact

Ranklets: orientation selective non-parametric features applied to face detection

Abstract

We introduce a family of multiscale, orientation-selective, non-parametric features ("ranklets'') modelled on Haar wavelets. We clarify their relation to the Wilcoxon rank-sum test and the rank transform and provide an efficient scheme for computation based on the Mann-Whitney statistics. Finally, we show that ranklets outperform other rank features, Haar wavelets, SNoW and linear SVMs (based on independently published results) in face detection experiments over the 24'045 test images in the MIT-CBCL database.

Full text

Click here to download the full text of the paper.

Cite

6. Smeraldi, "Ranklets: orientation selective non-parametric features applied to face detection", 2002 International Conference on Pattern Recognition, Quebec City, QC, Canada, 2002, pp. 379-382 vol.3, doi: 10.1109/ICPR.2002.1047924.

BibTex

@InProceedings{smeraldi2002ranklets,
author = {F. Smeraldi},
title = {Ranklets: orientation selective non-parametric features applied to face detection},
booktitle = {Proc. of the 2002 International Conference on Pattern Recognition, Quebec QC, Canada},
pages = {379--382},
year = {2002},
volume = {3},
month = {August},
doi={10.1109/ICPR.2002.1047924}
}