Fig 4 - available via license: CC BY-ND
Content may be subject to copyright.
Download
View publication
A randomly chosen subset of the MB-LBP features E. Choosing weak classifiers For each MB-LBP feature, we adopt multi-branch tree as weak classifiers. The multi-branch tree totally has 256 branches, and each branch corresponds to a certain discrete value of MB-LBP features. The weak classifier can be defined as:

A randomly chosen subset of the MB-LBP features E. Choosing weak classifiers For each MB-LBP feature, we adopt multi-branch tree as weak classifiers. The multi-branch tree totally has 256 branches, and each branch corresponds to a certain discrete value of MB-LBP features. The weak classifier can be defined as:

Source publication
Fig. 1. An illustration of the great challenges of eye localization...
Fig. 2. The geometry of eye training samples: a-larger scale;...
Fig. 4. A randomly chosen subset of the MB-LBP features E. Choosing...
Fig. 6. Results of eye center localization for the BioID database
Eye center localization on a facial image based on multi-block local binary patterns
Article
Full-text available
  • Jun 2015
In this paper an eye center localization algorithm based on multi-block local binary patterns is described. Performance of the suggested algorithm is compared to another methods based on Bayesian approach and image gradients. Visual examples of eye center localization results are provided.
Cite
Download full-text

Context in source publication

Context 1
... the center rectangle, are rectangles (see Fig. 3 for an tterns may be obtained for a m are presented in Fig. 4. ...

Citations

Improving the eyes localization algorithm for face recognition systems
Conference Paper
  • Sep 2015
An eye center localization algorithm based on multi-block local binary patterns is described. Performance of the suggested algorithm is compared to other popular approaches on standard BioID and FERET image databases. Additional test on images with two types of distortions: Additive white Gaussian noise and JPEG compression is considered. The impact of eye center localization accuracy on face recognition rate is analyzed.
View