Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
½Ç½Ã°£ ¿µ»ó ÃÊÇØ»óµµ º¹¿øÀ» À§ÇÑ È¿À²ÀûÀÎ ½Å°æ¸Á ±¸Á¶ ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Study of Efficient Network Structure for Real-time Image Super-Resolution |
ÀúÀÚ(Author) |
Á¤¿ìÁø
ÇѺ¹±Ô
À̵¿¼®
ÃÖº´ÀÎ
¹®¿µ½Ä
Woojin Jeong
Bok Gyu Han
Dong Seok Lee
Byung In Choi
Young Shik Moon
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 19 NO. 04 PP. 0045 ~ 0052 (2018. 08) |
Çѱ۳»¿ë (Korean Abstract) |
´ÜÀÏ ¿µ»ó ÃÊÇػ󵵴 ÇϳªÀÇ ÀúÇØ»óµµ ¿µ»ó¿¡¼ °íÇØ»óµµ ¿µ»óÀ» º¹¿øÇÏ´Â °úÁ¤ÀÌ´Ù. ÃÖ±Ù ½ÉÃþ½Å°æ¸ÁÀ» Àû¿ëÇÑ ÃÊÇØ»óµµ ±â¹ýÀÌ ÁÁÀº ¼º°ú¸¦ ³ªÅ¸³»°í ÀÖ´Ù. º» ³í¹®¿¡¼´Â ±âÁ¸ÀÇ ½ÉÃþ½Å°æ¸Á ±â¹Ý ÃÊÇØ»óµµ º¹¿ø ±â¹ýº¸´Ù ¼Óµµ¿Í ¼º´ÉÀ» °³¼±ÇÑ ½Å°æ¸Á ±¸Á¶¸¦ Á¦¾ÈÇÑ´Ù. À̸¦ À§ÇØ ±âÁ¸ ±â¹ýÀÇ ´ÜÁ¡À» ºÐ¼®ÇÏ°í ÇØ°áÃ¥À» Á¦½ÃÇÑ´Ù. Á¦¾ÈÇÏ´Â ¹æ¹ýÀº ±âÁ¸ ±â¹ýÀÇ 5´Ü°è¸¦ 3´Ü°è·Î ÁÙ¿© È¿À²¼ºÀ» ³ô¿´À¸¸ç, ³×Æ®¿öÅ©ÀÇ Æø°ú ±íÀÌ¿¡ ´ëÇÑ ½ÇÇèÀ» ÅëÇØ °¡Àå È¿À²ÀûÀÎ ½Å°æ¸Á ±¸Á¶¸¦ ¿¬±¸ÇÏ¿´´Ù. Á¦¾ÈÇÏ´Â ¹æ¹ýÀÇ ¼º´É°ú ¼Óµµ¸¦ ¾Ë¾Æº¸±â À§ÇØ ºñ±³ ½ÇÇèÀ» ÁøÇàÇÏ¿´´Ù. Á¦¾ÈÇÏ´Â ¹æ¹ýÀº 1024¡¿1024 ¿µ»óÀ» ÃÊ´ç 148Àå º¹¿øÇÏ´Â ¼Óµµ¸¦ ³ªÅ¸³ÂÀ¸¸ç, 4°¡Áö µ¥ÀÌÅÍ¿¡ ´ëÇØ ±âÁ¸ ¹æ¹ýº¸´Ù º¹¿ø ¼º´ÉÀÌ ¿ì¼öÇÏ¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
A single-image super-resolution is a process of restoring a high-resolution image from a low-resolution image. Recently, the super-resolution using the deep neural network has shown good results. In this paper, we propose a neural network structure that improves speed and performance over conventional neural network based super-resolution methods. To do this, we analyze the conventional neural network based super-resolution methods and propose solutions. The proposed method reduce the 5 stages of the conventional method to 3 stages. Then we have studied the optimal width and depth by experimenting on the width and depth of the network. Experimental results have shown that the proposed method improves the disadvantages of the conventional methods. The proposed neural network structure showed superior performance and speed than the conventional method.
|
Å°¿öµå(Keyword) |
¿µ»ó ÃÊÇØ»óµµ º¹¿ø
½ÉÃþ½Å°æ¸Á
ÄÁº¼·ç¼Ç ½Å°æ¸Á
Image super-resolution
deep neural network
Convolutional Neural Network
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|