Comparison of confidence measures for face recognition

Author

Eickeler, Stefan ; Jabs, Mirco ; Rigoll, Gerhard

Author_Institution

Dept. of Comput. Sci., Duisburg Univ., Germany

fYear

2000

fDate

2000

Firstpage

257

Lastpage

262

Abstract

This paper compares different confidence measures for the results of statistical face recognition systems. The main applications of a confidence measure are rejection of unknown people and the detection of recognition errors. Some of the confidence measures are based on the posterior probability and some on the ranking of the recognition results. The posterior probability is calculated by applying Bayes´ rule with different ways to approximate the unconditional likelihood. The confidence measure based on the ranking is a new method. Experiments to evaluate the confidence measures are carried out on a pseudo 2D hidden Markov model-based face recognition system and the Bochum face database

Keywords

Bayes methods; face recognition; hidden Markov models; probability; statistical analysis; Bayes rule; Bochum face database; confidence measures; posterior probability; pseudo 2D hidden Markov model; recognition error detection; statistical face recognition systems; unconditional likelihood approximation; unknown people rejection; Computer science; Discrete cosine transforms; Electric variables measurement; Face detection; Face recognition; Hidden Markov models; Image sampling; Probability; Speech recognition; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Automatic Face and Gesture Recognition, 2000. Proceedings. Fourth IEEE International Conference on

Conference_Location

Grenoble

Print_ISBN

0-7695-0580-5

Type

conf

DOI

10.1109/AFGR.2000.840644

Filename

840644