Comparing landmarking methods for face recognition


Share/Save/Bookmark

Beumer, G.M. and Tao, Q. and Bazen, A.M. and Veldhuis, R.N.J. (2005) Comparing landmarking methods for face recognition. In: ProRISC 2005, 16th Workshop on Circuits, Systems and Signal Processing, 17-18 November 2005, Veldhoven, The Netherlands (pp. pp. 594-597).

open access
[img]
Preview
PDF
138kB
Abstract:Good registration (alignment to a reference) is essential for accurate face recognition. We use the locations of facial features (eyes, nose, mouth, etc) as landmarks for registration. Two landmarking methods are explored and compared: (1) the Most Likely-Landmark Locator (MLLL), based on maximizing the likelihood ratio [1], and (2) Viola-Jones detection [2]. Further, a landmark-correction method based on projection into a subspace is introduced. Both landmarking methods have been trained on the landmarked images in the BioID database [3]. The MLLL has been trained for locating 17 landmarks and the Viola-Jones method for 5 landmarks. The localization error and effects on the equal-error rate (EER) have been measured. In these experiments ground- truth data has been used as a reference. The results are described as follows:
1. The localization errors obtained on the FRGC database are 4.2, 8.6 and 4.6 pixels for the Viola-Jones, the MLLL, and the MLLL after landmark correction, respectively. The inter-eye distance of the reference face is 100 pixels. The MLLL with landmark correction scores best in the verification experiment.
2. Using more landmarks decreases the average localization error and the EER.
Item Type:Conference or Workshop Item
Copyright:© STW, Technology Foundation 2005
Faculty:
Electrical Engineering, Mathematics and Computer Science (EEMCS)
Research Group:
Link to this item:http://purl.utwente.nl/publications/59557
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page

Metis ID: 226835