An Improved Multi-target Tracking Method Integrating Visual Tracking Mechanism
Konferenz: ICMLCA 2021 - 2nd International Conference on Machine Learning and Computer Application
17.12.2021 - 19.12.2021 in Shenyang, China
Tagungsband: ICMLCA 2021
Seiten: 5Sprache: EnglischTyp: PDF
Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Autoren:
Yang, Dawei; Qi, Yan; Liu, Liping (School of Information Technology and Engineering, Shenyang Ligong University, Shenyang, China)
Wei, Chenyang (Lijin Senior High School, Dongying, China)
Inhalt:
In this paper, to better solve the problem of frequent ID changes caused by false detection, missed detection and target occlusion in multi-target tracking, an improved multi-target tracking method integrating visual tracking mechanism is proposed. GOTURN algorithm integrated in the tracking method plays the role of visual tracking mechanism. Comparative experimental results between the improved method and the original method demonstrate the effectiveness of the proposed method.