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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) °¡»ó Ä¡¾Æ À̹ÌÁö »ý¼ºÀ» À§ÇÑ µö·¯´× ¸ðµ¨ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) A Study on Deep Learning Model for Virtual Tooth Image Generation
ÀúÀÚ(Author) Á¤¼ö¿¬   ¹èÀºÁ¤   ÀåÇö¼ö   ³ª½ÂÁÖ   ÀÓ¼±¿µ   Soo-Yeon Jeong   Eun-Jeong Bae   Jang Hyun Soo   SeongJu Na   Sun-Young Ihm  
¿ø¹®¼ö·Ïó(Citation) VOL 38 NO. 03 PP. 0073 ~ 0082 (2022. 12)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù µö·¯´× ±â¼úÀÇ ¹ßÀüÀ¸·Î ´Ù¾çÇÑ ºÐ¾ß¿¡¼­ µö·¯´× ¸ðµ¨ÀÌ È°¿ëµÇ°í ÀÖ´Ù. ±× Áß GANÀº »ý¼ºÀÚ¿Í ÆǺ°ÀÚ ¸ðµ¨ÀÌ ¼­·Î °æÀïÇÏ¿© µ¥ÀÌÅ͸¦ »ý¼ºÇÏ´Â µö·¯´× ¸ðµ¨À̸ç, °¡»óÀÇ À̹ÌÁö¸¦ »ý¼ºÇϴµ¥ ÁÖ·Î È°¿ëµÇ°í ÀÖ´Ù. ÃÖ±Ù Ä¡°ú º¸Ã¶¹°À» Á¦ÀÛÇÏ´Â °úÁ¤¿¡¼­ ÀΰøÁö´É°ú °ü·ÃµÈ ¿¬±¸°¡ È°¹ßÈ÷ ÁøÇàµÇ°í ÀÖÀ¸¸ç ƯÈ÷ GANÀº °¡»óÄ¡ ¾Æ¸¦ »ý¼ºÇÒ ¼ö ÀÖ´Ù´Â ÀåÁ¡À¸·Î Ä¡°úº¸Ã¶¹° Á¦ÀÛ¿¡ µö·¯´× ±â¼úÀ» Àû¿ëÇÏ°í ÀÖ´Ù. ÇÏÁö¸¸, ÃÖ±Ù µµÀÔµÈ ±¸°­½ºÄ³³Ê¸¦ ÅëÇØ ½ºÄµ µÈ À̹ÌÁö¿¡ ´ëÇÑ ¿¬±¸´Â ¹ÌºñÇÑ ¼öÁØÀ̸ç, ÇнÀÀ» Çϱ⿡´Â À̹ÌÁö ¼ö°¡ ºÎÁ·ÇÑ ½ÇÁ¤ÀÌ´Ù. µû¶ó¼­ º» ¿¬±¸¿¡¼­´Â GAN ±â¹ÝÀÇ µö·¯´× ¸ðµ¨À» È°¿ëÇÏ¿© °¡»óÀÇ ±¸°­ À̹ÌÁö ¹× »ó½Ç Ä¡¾Æ ¿µ¿ªÀ» ÀÎÆäÀÎÆÃÇÏ´Â µö·¯´× ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. Á¦¾È ¹æ¹ýÀº ±¸°­½ºÄ³³Ê·Î ½ºÄµ µÈ ±¸°­ À̹ÌÁö¿Í Pix2Pix¸¦ ÅëÇØ »ý¼ºÇÑ À̹ÌÁö¸¦ ÀÎÆäÀÎÆà ±â¹ýÀÎ CR-Fill¿¡ È°¿ëÇÑ´Ù. CR-FillÀº »ó½Ç Ä¡¾Æ ¿µ¿ª¿¡ ÀǹÌÀÖ´Â Ä¡¾Æ¸¦ »ý¼ºÇÒ ¼ö ÀÖµµ·Ï ÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
With the recent development of deep learning technology, deep learning models are being used in various fields. Among them, GAN is a deep learning model in which generator and discriminator models compete with each other to generate data, and is mainly used to create virtual images. Recently, research related to artificial intelligence is being actively conducted in the process of manufacturing dental prosthesis, and in particular, GAN is applying deep learning technology to manufacturing dental prosthesis with the advantage of being able to create virtual teeth. However, research on images scanned through the recently introduced intraoral scanner is insufficient, and the number of images is insufficient for learning. Therefore, in this study, we propose a deep learning model that inpaints a virtual oral image and missing tooth area using a GAN-based deep learning model. The proposed method utilizes the oral image scanned with an intraoral scanner and the image generated through pix2pix for CR-Fill, an inpainting technique. CR-Fill makes it possible to create meaningful teeth in the area of missing teeth.
Å°¿öµå(Keyword) °¡»ó Ä¡¾Æ À̹ÌÁö   À̹ÌÁö ÀÎÆäÀÎÆà  GAN   CR-Fill   Pix2Pix   Virtual Tooth Image   Image Inpainting  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå