KOFACEKorea Foundation for Asian Culture Exchange
Copyright 1988-2018 AcronymFinder.com, All rights reserved.
References in periodicals archive ?
In Table 1, we describe the datasets for all experiments conducted in the KoFace, Yale, and ORL database.
A large dataset used in this process includes 40 face subjects in the KoFace database.
In this section, we compare our proposed approach using visual observation confidence and the GMM baseline approach, and we examine the face recognition performance based on the Yale, ORL, and KoFace databases.
While the increasing rates are only 1.33% and 1% in the cases of the Y ale and ORL databases, respectively, the increasing rate is particularly high at 22.66% on the KoFace database in comparison with the GMM baseline rate.
Error Reduction is calculated by the difference of error rate in baseline method and error rate in proposed method over the error rate in baseline method.The figure shows a high error reduction of 89.46% and 100% in the KoFace and Yale databases, respectively.
From the KoFace database, we also conduct more comparison results between some other illumination techniques with our VOC model.