《計算機應用研究》|Application Research of Computers

基于視覺的人體行為識別算法研究綜述

Survey of human action recognition algorithm based on vision

免費全文下載 (已被下載 次)  
獲取PDF全文
作者 陳煜平,邱衛根
機構 廣東工業大學 計算機學院,廣州 510006
統計 摘要被查看 次,已被下載
文章編號 1001-3695(2019)07-002-1927-08
DOI 10.19734/j.issn.1001-3695.2018.04.0259
摘要 主要講述人體行為識別的基礎流程,歸納了人體行為識別常用的數據集,總結了時域分割的發展現狀和常用的方法,講解了人體行為識別比較經典的方法,并歸納了人體行為識別最新、最熱的深度學習方法。引入了動作分割,再結合行為識別,能夠實現連續的人體行為識別,使得行為識別適用于實際場景,而不再是對經過人工剪輯好的單個視頻進行識別,這在實際應用中意義重大。
關鍵詞 人體行為識別; 數據集; 動作分割; 深度學習; 雙流網絡
基金項目 國家自然科學基金資助項目(61572142)
廣東省科技計劃資助項目(14ZK0180)
本文URL http://www.oirznw.live/article/01-2019-07-002.html
英文標題 Survey of human action recognition algorithm based on vision
作者英文名 Chen Yuping, Qiu Weigen
機構英文名 School of Computers,Guangdong University of Technology,Guangzhou 510006,China
英文摘要 This paper focused on action recognition and included data sets and motion segmentation. It mainly described the basic flow of human action recognition. And it summarized the commonly used data sets of human action recognition. Then it summarized the development status and common methods of time domain segmentation. Next it explained the classic algorithms of human action recognition. At last, it summarized the the-state-of-art deep learning methods of human action recognition. The introduction of action recognition combines with action segmentation made the action recognition applicable to the actual scene, which could achieves continuous recognition of human action. Meanwhile it was no longer recognize a single video that has been manually edited. This has very important reference value in practical applications.
英文關鍵詞 human action recognition; data set; motion segmentation; deep learning; two-stream network
參考文獻 查看稿件參考文獻
 
收稿日期 2018/4/30
修回日期 2018/6/11
頁碼 1927-1934
中圖分類號 TP391.41
文獻標志碼 A
水果机返水