立命館大学グローバルCOE　デジタルアーカイブ技術研究班

2010年3月23日

IEVC2010 workshop: Nice, France

 On March 5-7 2010, The IIEEJ Image Electronics and Visual Computing Workshop 2010 (IEVC2010) was held in Nice, France as the 2nd international academic event of Image Electronics Engineers of Japan (IIEEJ). This workshop aims to exchanging the knowledge of the latest researches in all areas of image electronics and visual computing. 

Within this three day of the workshop, I have an opportunity to discuss with and listen to many researcher works. Most of the researchers are from Japan since this workshop was organized by IEEJ. There are 110 participates and more than 80 presenters including oral and poster presentations. Among the presenters, three oral presentations including myself were done by researchers from Ritsumeikan University.

My presentation topic is "Improved Segmentation of Motion Capture Data Using SMOTE". In this presentation, I introduced a new method for increasing the accuracy of an automatic segmentation of a long dance motion into unit motions. SMOTE, Synthetic Minority Over-sampling Technique, increases the classification accuracy of using neural networks in case of the training set including the minority class.

In this workshop, I have a great experience not only to conduct my ongoing researches but also to create a social network from professors, researches, Ph.D. students. Moreover, the experience in France is remarkable. Nice is a wonderful French Riviera city with a gorgeous coastline, beautiful traditional style houses and very modern city. This was my first visited to European country which gave me the opportunity to experience foods, cultures, life styles and many things which I never experience before.

I would like to thank all people in GCOE program that support my research and Professor Kozabura Hachimura, who is my supervisor.    

P.S If anyone plans for shopping in France, please beware that your favorite's shops maybe close on Sunday since most of the shop is close on Sunday.

Worawat, CHOENSAWAT

by kh | 記事詳細 | コメント[0] | トラックバック[0] [編集]

コメントする