DocumentCode :
2345817
Title :
Efficient evaluation of classification and recognition systems
Author :
Micheals, Ross J. ; Boult, Terrance E.
Author_Institution :
Comput. Sci. & Eng. Dept., Lehigh Univ., Bethlehem, PA, USA
Volume :
1
fYear :
2001
fDate :
2001
Abstract :
In this paper, a new framework for evaluating a variety of computer vision systems and components is introduced. This framework is particularly well suited for domains such as classification or recognition systems, where blind application of the i.i.d. assumption would reduce an evaluation´s accuracy, such as with classification or recognition systems. With few exceptions, most previous work on vision system evaluation does not include confidence intervals, since they are difficult to calculate, and are often coupled with strict requirements. We show how a set of previously overlooked replicate statistics tools can be used to obtain tighter confidence intervals of evaluation estimates while simultaneously reducing the amount of data and computation required to reach such sound evaluatory conclusions. In the included application of the new methodology, the well-known FERET face recognition system evaluation is extended to incorporate standard errors and confidence intervals.
Keywords :
computer vision; face recognition; image classification; image recognition; FERET face recognition system; computer vision systems; confidence intervals; image classification; image recognition; replicate statistics tools; Application software; Computer errors; Computer science; Computer vision; Face detection; Face recognition; Layout; Machine vision; Statistics; Terminology;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition, 2001. CVPR 2001. Proceedings of the 2001 IEEE Computer Society Conference on
ISSN :
1063-6919
Print_ISBN :
0-7695-1272-0
Type :
conf
DOI :
10.1109/CVPR.2001.990455
Filename :
990455
Link To Document :
بازگشت