• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) Àû´ëÀû »ý¼º ½Å°æ¸ÁÀ» ÀÌ¿ëÇÑ ¾ó±¼ °¨Á¤ÀÎ½Ä µ¥ÀÌÅÍ Áõ°­
¿µ¹®Á¦¸ñ(English Title) Facial Emotion Recognition Data Augmentation using Generative Adversarial Network
ÀúÀÚ(Author) ±èÁø¿ë   Á¶±Ù½Ä   Jinyong Kim   Geunsik Jo  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 04 PP. 0398 ~ 0404 (2021. 04)
Çѱ۳»¿ë
(Korean Abstract)
ÄÄÇ»ÅÍ ºñÀüÀÇ ¾ó±¼ °¨Á¤ÀÎ½Ä ºÐ¾ß´Â µö·¯´×ÀÇ ´Ù¾çÇÑ ½Å°æ¸ÁÀ» ÅëÇØ ÃÖ±Ù ÀǹÌÀÖ´Â Çຸ¸¦ º¸ÀÌ°í ÀÖ´Ù. ±×·¯³ª ÁÖ¿äÇÏ°Ô »ç¿ëµÇ´Â µ¥ÀÌÅͼµéÀº ¡°Å¬·¡½º ºÒ±ÕÇü¡±À̶ó´Â ¹®Á¦¸¦ ¾È°í ÀÖ°í ÀÌ´Â µö·¯´× ¸ðµ¨ÀÇ Á¤È®µµ¸¦ Ç϶ô½ÃÅ°´Â ¿äÀÎÀÌ µÈ´Ù. ±×·¯¹Ç·Î Ŭ·¡½º ºÒ±ÕÇüÀ̶ó´Â ¹®Á¦¸¦ ÇؼÒÇϱâ À§ÇÑ ¿¬±¸µé ÀÌ È°¹ßÇÏ°Ô ÁøÇàµÇ°í ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ¾ó±¼ °¨Á¤ÀÎ½Ä µ¥ÀÌÅͼÂÀ¸·Î »ç¿ëµÇ´Â FER2013, RAF_single µ¥ÀÌÅͼÂÀÇ Å¬·¡½º ºÒ±ÕÇüÀ» ÇؼÒÇϱâ À§ÇØ Àû´ëÀû »ý¼º ½Å°æ¸ÁÀ» ÀÌ¿ëÇÑ ¾ó±¼ °¨Á¤ÀÎ½Ä µ¥ÀÌÅÍ Áõ°­ ¸ðµ¨ÀÎ ¡°RDGAN¡±À» Á¦¾ÈÇÑ´Ù. RDGANÀº ±âÁ¸ À̹ÌÁö °£ º¯È¯À» À§ÇÑ Àû´ëÀû »ý¼º ½Å°æ¸ÁÀ» ¹ÙÅÁÀ¸·Î Ç¥Çö ÆǺ°ÀÚ¸¦ Ãß°¡ÇÏ¿© ±âÁ¸ ¿¬±¸º¸´Ù Ŭ·¡½º¿¡ ÀûÇÕÇÑ À̹ÌÁö¸¦ »ý¼º ¹× º¯È¯ÇÏ´Â ³×Æ®¿öÅ©ÀÌ´Ù. RDGAN À¸·Î Áõ°­µÈ µ¥ÀÌÅͼÂÀº µ¥ÀÌÅÍ Áõ°­À» ÇÏÁö ¾ÊÀº µ¥ÀÌÅͼ°ú ºñ±³ÇÏ¿© FER2013°ú RAF_single¿¡¼­ °¢°¢ Æò±Õ 4.805%p, 0.857%pÀÇ ¼º´É Çâ»óÀ» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
The facial emotion recognition field of computer vision has recently been identified to demonstrate meaningful results through various neural networks. However, the major datasets of facial emotion recognition have the problem of "class imbalance," which is a factor that degrades the accuracy of deep learning models. Therefore, numerous studies have been actively conducted to solve the problem of class imbalance. In this paper, we propose "RDGAN," a facial emotion recognition data augmentation model that uses a GAN to solve the class imbalance of the FER2013 and RAF_single that are used as facial emotion recognition datasets. RDGAN is a network that generates images suitable for classes by adding expression discriminators based on the image-to-image translation model between the existing images as compared to the prevailing studies. The dataset augmented with RDGAN showed an average performance improvement of 4.805%p and 0.857%p in FER2013 and RAF_single, respectively, compared to the dataset without data augmentation.
Å°¿öµå(Keyword) Àû´ëÀû »ý¼º ½Å°æ¸Á   µ¥ÀÌÅÍ Áõ°­   ¾ó±¼ °¨Á¤ ÀνĠ  À̹ÌÁö °£ º¯È¯   generative adversarial network   data augmentation   facial emotion recognition   imageto-image translation  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå