Abstract: Although deep learning algorithm are widely used in computer vision, in consideration of the efficiency, most object trackers in video still uses traditional methods. In this paper, we propose a long-term tracking system based on deep learning algorithm, which could track fast moving objects and judge whether the tracking target is lost and retrieve it. To improve the performance of tracking fast moving objects, we proposed a motion model for the existing tracker(GOTURN). To meet the long-term tracking requirements, a detector, which can judge the lost target and retrieve it, is combined with the tracking system. We show that by aligned with the motion model and detector, the tracking system performance on the test dataset improves obviously.
Publication Year: 2018
Publication Date: 2018-01-05
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot