http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
오늘 본 자료
LSTM Network with Tracking Association for Multi Object Tracking
Xurshedjon Farhodov,Kwang-Seok Moon(문광석),Seoung-Khun Kwon(권성근),Suk-Hwan Lee(이석환),Ki-Ryong Kwon(권기룡) 대한전자공학회 2020 대한전자공학회 학술대회 Vol.2020 No.8
Today’s new AI technological development requires much more efficient and accurate controlling unit like object tracking than before. Multi object tracking tasks is one of the essential parts of the controlling systems at all. In this work we presented RNN LSTM model-based predicting with tracking association by using opensource datasets to test our system with different environmental video sequences. The learning process of training model goes offline to get trained model for applying tracking process to get more accurate and higher visual qualitive results.