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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Current Result Document : 10 / 91 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) RGB-D Á¤º¸¸¦ ÀÌ¿ëÇÑ °´Ã¼ ŽÁö ±â¹ÝÀÇ ½Åü Å°Æ÷ÀÎÆ® °ËÃâ ¹æ¹ý
¿µ¹®Á¦¸ñ(English Title) A Method for Body Keypoint Localization based on Object Detection using the RGB-D information
ÀúÀÚ(Author) ¹Ú¼­Èñ   ÀüÁØö   Seohee Park   Junchul Chun     
¿ø¹®¼ö·Ïó(Citation) VOL 18 NO. 06 PP. 0085 ~ 0092 (2017. 12)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù ¿µ»ó°¨½Ã ºÐ¾ß¿¡¼­´Â ¿µ»ó¿¡¼­ ¿òÁ÷ÀÌ´Â »ç¶÷À» ŽÁöÇÏ°í, ŽÁöµÈ »ç¶÷ÀÇ ÇàÀ§¸¦ ºÐ¼®ÇÏ´Â ¹æ½Ä¿¡ µö·¯´× ±â¹Ý ÇнÀ¹æ¹ýÀÌ Àû¿ëµÇ±â ½ÃÀÛÇß´Ù. ÀÌ·¯ÇÑ Áö´ÉÇü ¿µ»óºÐ¼® ±â¼úÀ» Àû¿ëÇÒ ¼ö ÀÖ´Â ºÐ¾ß Áß ÇϳªÀÎ Àΰ£ ÇàÀ§ ÀνÄÀº °´Ã¼¸¦ ŽÁöÇÏ°í ŽÁöµÈ °´Ã¼ÀÇ ÇàÀ§¸¦ ÀνÄÇϱâ À§ÇØ ½Åü Å°Æ÷ÀÎÆ®¸¦ °ËÃâ ÇÏ´Â °úÁ¤À» °ÅÄ¡°Ô µÈ´Ù. º» ³í¹®¿¡¼­´Â RGB-D Á¤º¸¸¦ ÀÌ¿ëÇÑ °´Ã¼ ŽÁö ±â¹ÝÀÇ ½Åü Å°Æ÷ÀÎÆ® °ËÃâ ¹æ¹ýÀ» Á¦½ÃÇÑ´Ù. ¸ÕÀú, µÎ ´ëÀÇ Ä«¸Þ¶ó·Î »ý¼ºµÈ »ö»óÁ¤º¸¿Í ±íÀÌÁ¤º¸¸¦ ÀÌ¿ëÇÏ¿© À̵¿ÇÏ´Â °´Ã¼¸¦ ¹è°æÀ¸·ÎºÎÅÍ ºÐÇÒÇÏ¿© ŽÁöÇÑ´Ù. RGB-D Á¤º¸¸¦ ÀÌ¿ëÇÏ¿© ŽÁöµÈ °´Ã¼ÀÇ ¿µ¿ªÀ» ÀçÁ¶Á¤ÇÏ¿© »ý¼ºµÈ ÀÔ·Â µ¥ÀÌÅ͸¦ ÇÑ »ç¶÷ÀÇ ÀÚ¼¼ ÃßÁ¤À» À§ÇÑ Convolutional Pose Machines(CPM)¿¡ Àû¿ëÇÑ´Ù. CPMÀ» ÀÌ¿ëÇÏ¿© ÇÑ »ç¶÷´ç 14°³ÀÇ ½ÅüºÎÀ§¿¡ ´ëÇÑ ½Å³ä Áöµµ(Belief Map)¸¦ »ý¼ºÇÏ°í, ½Å³ä Áöµµ¸¦ ±â¹ÝÀ¸·Î ½Åü Å°Æ÷ÀÎÆ®¸¦ °ËÃâÇÑ´Ù. ÀÌ¿Í °°Àº ¹æ¹ýÀº Å°Æ÷ÀÎÆ®¸¦ °ËÃâÇÒ °´Ã¼¿¡ ´ëÇÑ Á¤È®ÇÑ ¿µ¿ªÀ» Á¦°øÇÏ°Ô µÇ¸ç, °³º°ÀûÀÎ ½Åü Å°Æ÷ÀÎÆ®ÀÇ °ËÃâÀ» ÅëÇÏ¿© ´ÜÀÏ ½Åü Å°Æ÷ÀÎÆ® °ËÃâ¿¡¼­ ´ÙÁß ½Åü Å°Æ÷ÀÎÆ® °ËÃâ·Î È®Àå ÇÒ ¼ö ÀÖ´Ù. ÇâÈÄ, °ËÃâµÈ Å°Æ÷ÀÎÆ®¸¦ ÀÌ¿ëÇÏ¿© Àΰ£ ÀÚ¼¼ ÃßÁ¤À» À§ÇÑ ¸ðµ¨À» »ý¼ºÇÒ ¼ö ÀÖÀ¸¸ç Àΰ£ ÇàÀ§ ÀÎ½Ä ºÐ¾ß¿¡ ±â¿© ÇÒ ¼ö ÀÖ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recently, in the field of video surveillance, a Deep Learning based learning method has been applied to a method of detecting a moving person in a video and analyzing the behavior of a detected person. The human activity recognition, which is one of the fields this intelligent image analysis technology, detects the object and goes through the process of detecting the body keypoint to recognize the behavior of the detected object. In this paper, we propose a method for Body Keypoint Localization based on Object Detection using RGB-D information. First, the moving object is segmented and detected from the background using color information and depth information generated by the two cameras. The input image generated by rescaling the detected object region using RGB-D information is applied to Convolutional Pose Machines for one person's pose estimation. CPM are used to generate Belief Maps for 14 body parts per person and to detect body keypoints based on Belief Maps. This method provides an accurate region for objects to detect keypoints an can be extended from single Body Keypoint Localization to multiple Body Keypoint Localization through the integration of individual Body Keypoint Localization. In the future, it is possible to generate a model for human pose estimation using the detected keypoints and contribute to the field of human activity recognition.
Å°¿öµå(Keyword) ¿µ»ó°¨½Ã   °´Ã¼ ŽÁö   ½Åü Å°Æ÷ÀÎÆ® °ËÃâ   ÄÁº¼·ç¼Å³Î Æ÷Áî ¸Ó½Å   ½Å³ä Áöµµ   Àΰ£ ÇàÀ§ ÀνĠ  Video Surveillance   Object Detection   Body Keypoint Localization   Convolutional Pose Machines   Belief Map   Human Activity Recognition   ¿µ»ó°¨½Ã   °´Ã¼ ŽÁö   ½Åü Å°Æ÷ÀÎÆ® °ËÃâ   ÄÁº¼·ç¼Å³Î Æ÷Áî ¸Ó½Å   ½Å³ä Áöµµ   Àΰ£ ÇàÀ§ ÀνĠ  Video Surveillance   Object Detection   Body Keypoint Localization   Convolutional Pose Machines   Belief Map   Human Activity Recognition  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå