ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Light-weighted Mobile-Net based Human Pose Estimation for AR Service
Cited - time in scopus Share share facebook twitter linkedin kakaostory
Authors
Byung-Gyu Lee, Ju Young Kim, Sung-Uk Jung
Issue Date
2019-12
Citation
International Conference on Advanced Computing, Communication, and Information Sciences (ICACCI) 2019, pp.14-15
Language
English
Type
Conference Paper
Abstract
As deep neural network technology has been utilized in the field of image recognition, it is necessary to implement the light-weighted network to apply the AR service. In the sense, we propose a new method to estimate the human pose using the modified mobile-net in the AR service. The contribution of this paper is as following. i) the modified light-weighted network is applied to the AR service. ii) the human pose is extracted and analyzed in real time. By the properties, this system can be useful for augmented reality and educational purpose.
KSP Keywords
Augmented reality(AR), Deep neural network(DNN), Human Pose estimation, Image recognition, Network Technology, Real-time, neural network(NN), new method, weighted network