• 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¸Ã³¸®ÇÐȸ ³í¹®Áö > Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) TensorRT¿Í SSD¸¦ ÀÌ¿ëÇÑ ½Ç½Ã°£ ¾ó±¼ °ËÃâ¹æ¹ý
¿µ¹®Á¦¸ñ(English Title) Real Time Face detection Method Using TensorRT and SSD
ÀúÀÚ(Author) À¯Çýºó   ¹Ú¸í¼÷   ±è»óÈÆ   Hye-Bin Yoo   Myeong-Suk Park   Sang-Hoon Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 09 NO. 10 PP. 0323 ~ 0328 (2020. 10)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù¿¡´Â µö·¯´× ±â¼úÀ» ÀÌ¿ëÇÏ¿© ¹°Ã¼ °ËÃâ ¹× ÀνĿ¡¼­ ¼º´ÉÀÌ Å©°Ô Çâ»óµÇ´Â »õ·Î¿î Á¢±Ù¹æ¹ýµéÀÌ ºü¸£°Ô Á¦¾ÈµÇ°í ÀÖ´Ù. °´Ã¼, ƯÈ÷ ¾ó±¼°´Ã¼ °ËÃâ¿¡ °üÇÑ ¿©·¯ ±â¹ý(Faster R-CNN, R-CNN, YOLO, SSD µî) Áß SSD´Â ´Ù¸¥ ±â¹ýµéº¸´Ù Á¤È®µµ¿Í ¼Óµµ¿¡¼­ ¿ì¼öÇÏ´Ù. µ¿½Ã¿¡ ¿©·¯ °´Ã¼ °ËÃâ ³×Æ®¿öÅ©µé(object detection network)µµ ½±°Ô ÀÌ¿ëÇÒ ¼ö ÀÖ´Ù. º» ³í¹®¿¡¼­´Â °´Ã¼ °ËÃâ ³×Æ®¿öÅ© Áß Mobilenet v2 network¸¦ ÀÌ¿ëÇÏ°í SSD¿Í °áÇÕÇÑ ¸ðµ¨À» ÈÆ·ÃÇÏ°í, TensorRT engineÀ» ÀÌ¿ëÇÏ¿© ±âÁ¸ÀÇ ¼º´Éº¸´Ù 4¹è ÀÌ»óÀÇ ¼Óµµ·Î °´Ã¼¸¦ °ËÃâÇÏ´Â ¹æ¹ý¿¡ ´ëÇØ Á¦¾ÈÇÏ°í ½ÇÇèÀ» ÅëÇØ ¼º´ÉÀ» °ËÁõÇÑ´Ù. Á¦¾ÈÇÑ ¹æ¹ýÀÇ ¼º´É °ËÁõÀ» À§ÇÑ ÀÀ¿ëÀ¸·Î ¾ó±¼°´Ã¼ °ËÃâ±â(facial object detector)¸¦ ¸¸µé¾î ´Ù¾çÇÑ »óȲ¿¡¼­ µ¿ÀÛ°ú ¼º´ÉÀ» ½ÇÇèÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Recently, new approaches that significantly improve performance in object detection and recognition using deep learning technology have been proposed quickly. Of the various techniques for object detection, especially facial object detection (Faster R-CNN, R-CNN, YOLO, SSD, etc), SSD is superior in accuracy and speed to other techniques. At the same time, multiple object detection networks are also readily available. In this paper, among object detection networks, Mobilenet v2 network is used, models combined with SSDs are trained, and methods for detecting objects at a rate of four times or more than conventional performance are proposed using TensorRT engine, and the performance is verified through experiments. Facial object detector was created as an application to verify the performance of the proposed method, and its behavior and performance were tested in various situations.
Å°¿öµå(Keyword) ÅÙ¼­Ç÷ο젠 ÅÙ¼­¾ËƼ   µö·¯´×   ¿¡½º¿¡½ºµð   °´Ã¼ °ËÃâ   Tensorflow   TensorRT   Deep Learning   SSD   Object Detection  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå