IPIU 2018 (Á¦30ȸ ¿µ»óó¸® ¹× ÀÌÇØ¿¡ °üÇÑ ¿öÅ©¼¥)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
High-Speed 2D Human Motion Analysis Using Multi-Exposure Imaging and Deep Learning |
¿µ¹®Á¦¸ñ(English Title) |
High-Speed 2D Human Motion Analysis Using Multi-Exposure Imaging and Deep Learning |
ÀúÀÚ(Author) |
Jong-Sung Kim
Myung-Gyu Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 30 NO. 01 PP. P3 ~ 0023 (2018. 02) |
Çѱ۳»¿ë (Korean Abstract) |
|
¿µ¹®³»¿ë (English Abstract) |
In this paper, a new high-speed 2D human motion analysis method based on multi-exposure imaging and deep learning is proposed. The proposed method captures 2D human motion images with a multi-exposure camera, which makes it possible to obtain a single multi-exposed image of consecutive high-speed human motion. Then, a deep learning-based 2D human motion technique is applied to the multi-exposed image to analyze the high-speed 2D human motion in real time. The effectiveness and efficiency of the proposed method was demonstrated with real experimental results.
|
Å°¿öµå(Keyword) |
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|