Combining Matching Scores in Identification Model

Authors:
Sergey Tulyakov;Venu Govindaraju
Affiliations:
State University of New York at Buffalo, USA;State University of New York at Buffalo, USA
Venue:
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Year:
2005

Citing 4
Cited 1

Person Identification Using Multiple Cues

IEEE Transactions on Pattern Analysis and Machine Intelligence
Using HMM Based Recognizers for Writer Identification and Verification

IWFHR '04 Proceedings of the Ninth International Workshop on Frontiers in Handwriting Recognition
Speaker background models for connected digit password speaker verification

ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 01
Models of large population recognition performance

CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition

Fusion in Multibiometric Identification Systems: What about the Missing Data?

ICB '09 Proceedings of the Third International Conference on Advances in Biometrics

Quantified Score

Hi-index	0.00

Visualization

Abstract

The paper discusses a problem of combining recognition scores for different classes produced by one recognizer during one recognition attempt. This problem arises in identifi- cation problems which we define as 1:N classification problems with big or variable N. By using artificial example we show that intuitive solution of making identification decision based solely on the best matching score is frequently suboptimal. Paper presents reasons for such behavior, and draws parallels with score normalization technique used in speaker identification. Two examples of real life applications illustrate the possible benefits of properly combining recognition scores.