A New Multimodal Database for Performance Evaluation in System Level

Mi-young Cho
1+, Young-Sook Jeong1, Hyun-suh Kim2, and Howard Kim2
 

1Electronics and Telecommunications Research Institute, Daejeon, Korea
{mycho, ysjeong}@etri.re.kr

2Color Technology Laboratory, Suwon, Korea
{hyun, howard}@colortechlab.com

 

Abstract

Multimodal technology has become an important area of research because it holds great potential for overcoming certain problems of traditional speech recognition and face recognition methods. In spite of a variety of multimodal applications, it is difficult to guarantee of performance of the applications due to insufficient test in system level. To evaluate of performance in multimodal system level, we construct a new multimodal database which is calibrated by reference color chart.

Keywords: multimodal DB, delta E

 

+: Corresponding author: Mi-young Cho
Intelligence and Robot System Research Section, Electronics and Telecommunications Research

Institute(ETRI), Daejeon, Republic of Korea, Tel: +82-42-860-6453

IT Convergence Practice (INPRA), Vol. 3, No. 3, pp. 11-17, September 2015 [pdf]