• 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 ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë

Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë

Current Result Document : 53 / 99 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) 3Â÷¿ø ±íÀÌ ÃßÁ¤ Á¤º¸¸¦ ÀÌ¿ëÇÑ ´Éµ¿Àû ¿Ü¾ç¸ðµ¨ ±â¹Ý 3Â÷¿ø ¾ó±¼ ¸ðµ¨ ÇÇÆà ±â¹ý
¿µ¹®Á¦¸ñ(English Title) 3D Face Model Fitting Method based on Active Appearance Model with 3D Depth Estimation
ÀúÀÚ(Author) ÁÖ¸íÈ£   °­ÇàºÀ   Myung-Ho Ju   Hang-Bong Kang  
¿ø¹®¼ö·Ïó(Citation) VOL 39 NO. 02 PP. 0109 ~ 0117 (2012. 02)
Çѱ۳»¿ë
(Korean Abstract)
¾ó±¼ÀÇ 3Â÷¿ø Á¤º¸¸¦ ÀνÄÇϱâ À§Çؼ­´Â 3D ½ºÄ³³Ê, ±íÀÌ Ä«¸Þ¶ó µî°ú °°Àº °í°¡ÀÇ Àåºñ°¡ ¿ä±¸µÈ´Ù. º» ³í¹®¿¡¼­´Â ¼Õ½±°Ô ȹµæ °¡´ÉÇÑ 2Â÷¿øÀÇ ¾ó±¼ ¿µ»óÀ» ÇнÀÇÏ¿© 3Â÷¿øÀ¸·Î º¯È­ÇÏ´Â »ç¿ëÀÚ ¾ó±¼ÀÇ Æ÷Áî (X, Y, ZÃà À̵¿ ¹× ȸÀü) ¹× ¾ó±¼ Ç¥Á¤ º¯È­¿¡ ´ëÇÑ 3Â÷¿ø ¾ó±¼ Á¤º¸ÀÇ È¿°úÀûÀÎ ÃßÀû, ÇÇÆÃ(Fitting) ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈµÈ ¹æ¹ýÀº ÇнÀµÈ 2Â÷¿øÀÇ ¾ó±¼ ÇüÅÂ¿Í 3Â÷¿øÀÇ ±íÀÌ Á¤º¸¸¦ °áÇÕÇÏ¿© ÀÔ·Â ¾ó±¼À» ÇÇÆÃÇϱ⠶§¹®¿¡ ÀÔ·Â ¾ó±¼ÀÇ 3Â÷¿ø ¾ó±¼ Æ÷Áî°¡ ÇнÀ ¿µ»ó¿¡ ÀÇÁ¸ÀûÀÌÁö ¾ÊÀ¸¸ç »çÀü ÇнÀµÇÁö ¾ÊÀº ¾ó±¼ Æ÷Áî¿¡ ´ëÇؼ­µµ È¿°úÀûÀÎ 3Â÷¿ø ¾ó±¼ ÇÇÆÃÀ» ¼öÇàÇÑ´Ù. º» ³í¹®¿¡¼­´Â »ç¿ëÀÚÀÇ ¾ó±¼ Ç¥Á¤º¯È­¸¦ ÇнÀÇϱâ À§ÇØ ¸ÕÀú »ç¿ëÀÚÀÇ Á¤¸é ¾ó±¼¿¡ ´ëÇÑ Ç¥Á¤ º¯È­¸¦ Active Appearance ModelsÀ» ÀÌ¿ëÇÏ¿© ÇнÀÇÑ´Ù. ±×¸®°í Á¤¸é, Ãø¸éÀÇ ¾ó±¼ Æ÷Á °®´Â 3ÀåÀÇ ¿µ»óÀ¸·ÎºÎÅÍ ¾ó±¼ÀÇ ±íÀÌ Á¤º¸¸¦ ÃßÁ¤ÇÏ°í ÃßÁ¤µÈ ±íÀÌ Á¤º¸¿Í ÇнÀµÈ AAMÀ» °áÇÕÇÏ¿© ¾ó±¼ÀÇ 3Â÷¿ø Á¤º¸¸¦ ÇÇÆÃÇÑ´Ù. ¶ÇÇÑ ¾ó±¼ÀÇ Æ÷Áî º¯È­´Â 3Â÷¿ø ȸÀüÀ» Æ÷ÇÔÇϱ⠶§¹®¿¡ Àڱ⠰¡¸²(Self-Occlusion)ÀÌ ¹ß»ýµÈ´Ù. º» ³í¹®¿¡¼­´Â ÀÌ·¯ÇÑ Àڱ⠰¡¸²ÀÌ ¹ß»ýµÈ ¿µ¿ª¿¡ ´ëÇØ Á¤±Ô ¾ó±¼(Normalized face)ÀÇ °¡ÁßÄ¡¸¦ ºÎ¿©ÇÔÀ¸·Î½á º¸´Ù Á¤È®ÇÑ 3Â÷¿ø ¾ó±¼ ÇÇÆÃÀ» ¼öÇàÇÑ´Ù. º» ³í¹®¿¡¼­ Á¦¾ÈµÈ ¹æ¹ýÀº ¾ó±¼ÀÇ ´Ù¾çÇÑ Ç¥Á¤ º¯È­¿Í ÇÔ²² 3Â÷¿øÀÇ ¾ó±¼ Æ÷Áî º¯È­¸¦ Æ÷ÇÔÇÑ ½ÇÇè ¿µ»óÀ» ÀÌ¿ëÇÏ¿© ±âÁ¸ÀÇ AAM ¹× View-based AAM¿¡ ºñÇØ Á¤È®ÇÑ ¾ó±¼ ÇÇÆÃÀ» ¼öÇàÇÒ ¼ö ÀÖÀ½À» º¸ÀδÙ.
¿µ¹®³»¿ë
(English Abstract)
Special cameras (e.g. 3D scanners or depth cameras) are needed to recognize 3D shape information from input faces. In this paper, we propose an efficient face fitting method which can fit various 3D faces including various poses (the rotation of X, Y axies) and facial expressions. Our method takes advantage of 2D Active Appearance Models (AAM) from 2D face images rather than the depth information measured by special cameras. The proposed method combines 2D face model with depth information, thereby the poses of the input faces are not depend on the training data. We construct an AAM for the variation of facial expressions. Then, we estimate depth information of each land-mark from frontal, side view images. By combining the estimated depth information with AAM, we can fit various 3D faces. Self-occlusions due to the 3D pose variation are also processed by the region weighting function on the normalized face at each frame. Our proposed method can fit on various face poses which are not trained. Our experimental results show that the proposed method can efficiently fit various faces better than the typical AAM and View-based AAM.
Å°¿öµå(Keyword) 3D ¾ó±¼   ¾ó±¼ ÇÇÆà  3D ¾ó±¼ ÇüÅ   3D Face   Face Fitting   3D Face Shape  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå