• 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¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document : 24 / 128 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) »ý¼ºÀû Àû´ë ³×Æ®¿öÅ©¸¦ ÀÌ¿ëÇÑ ±ÛÀÚ Æ¯¼ºÀÌ °í·ÁµÈ ÃÊÇØ»óµµ ¿µ»ó º¹¿ø
¿µ¹®Á¦¸ñ(English Title) Image Super-Resolution with Text Handling Via Generative Adversairal Network
ÀúÀÚ(Author) À¯ÀçÇÊ   ¼ÕÇü¼®   Á¶¼ºÇö   À̽¿렠 Jaepil Yu   Hyeongseok Son   Sunghyun Cho   Seungyong Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 24 NO. 08 PP. 0405 ~ 0409 (2018. 08)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®Àº µö·¯´× ¸ðµ¨ Áß ÇϳªÀÎ generative adversarial network (GAN)À» »ç¿ëÇÏ¿© ±ÛÀÚ Æ¯¼ºÀÌ °í·ÁµÈ ÃÊÇØ»óµµ ¿µ»ó º¹¿ø ¹æ¹ýÀ» Á¦½ÃÇÑ´Ù. ±âÁ¸ÀÇ ÃÊÇØ»óµµ ¿µ»ó º¹¿ø ¹æ¹ýÀº ÀϹÝÀûÀÎ ¿µ»ó¿¡ ´ëÇÑ Æ¯Â¡µéÀ» ÁÖ·Î ÇнÀÇϱ⠶§¹®¿¡ ±ÛÀÚ ¿µ¿ªÀÇ º¹¿ø¿¡ ´ëÇؼ­´Â ºÎÁ·ÇÑ ¼º´ÉÀ» º¸ÀδÙ. ±ÛÀÚ ¿µ»óÀÌ °¡Áö°í Àִ Ư¡Àº ÀÏ¹Ý ¿µ»óÀÇ Æ¯Â¡°ú ±¸ºÐµÇ¹Ç·Î À̸¦ º°µµ·Î ó¸®Çϱâ À§ÇÑ °úÁ¤ÀÌ ÇÊ¿äÇÏ´Ù. µû¶ó¼­ º» ³í¹®¿¡¼­´Â ±âÁ¸ÀÇ µ¥ÀÌÅͼ¿¡ ±ÛÀÚ¸¦ Ãß°¡ÇÏ°í, ÀÏ¹Ý ¿µ»ó¿¡ ´ëÇÑ ÇнÀ°ú ±ÛÀÚ ¿µ»ó¿¡ ´ëÇÑ ÇнÀÀ» ³ª´©¾î ¼öÇàÇÏ¿© ±ÛÀÚ ¿µ¿ª¿¡ ´ëÇØ °³¼±µÈ ÃÊÇØ»óµµ º¹¿ø ¹æ¹ýÀ» Á¦½ÃÇÑ´Ù. ½ÇÇè °á°ú¸¦ ÅëÇØ º» ³í¹®¿¡¼­ Á¦¾ÈÇÑ ¾Ë°í¸®ÁòÀÌ ±ÛÀÚ°¡ Æ÷ÇÔµÈ ¿µ»ó¿¡ ´ëÇÏ¿© º¹¿øÀÇ Ç°ÁúÀÌ Çâ»óµÇ´Â °ÍÀ» º¸ÀδÙ.
¿µ¹®³»¿ë
(English Abstract)
We propose a text-image super-resolution method using a generative adversarial network (GAN). Because previous super-resolution methods mainly learned properties of natural images, the quality of restored text regions in a super-resolution image is relatively low. The characteristics of text images and natural images are different, requiring an additional process to treat property of texts. We solve the text-image super-resolution problem by training natural images and text images separately. We add some texts to the dataset and use them for training the network to handle tex regions as well as natural-image regions. Experimental results show that the proposed network produces better text-image quality in the super-resolution results.
Å°¿öµå(Keyword) ¿µ»ó 󸮠  ÃÊÇØ»óµµ ¿µ»ó   µö·¯´×   »ý¼ºÀû Àû´ë ³×Æ®¿öÅ©   image processing   super-resolution   deep learning   generative adversarial network  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå