摘要
当前,基于目标跟踪技术的应用产品在教育、医学及军事等领域得到广泛应用。深度学习的崛起使得目标跟踪技术有了突破性进展,它冲破传统机器学习的束缚,创造性地提出许多鲁棒的跟踪模型。基于此,系统地研究深度跟踪技术,总结分析深度跟踪技术的发展现状和未来趋势。
At present,application products based on target tracking technology are widely used in education,medicine,military and other fields.The rise of deep learning makes a breakthrough in target tracking technology.It breaks through the shackles of traditional machine learning and creatively proposes many robust tracking models.Based on this,this paper systematically studies the depth tracking technology,summarizes and analyzes the development status and future trend of depth tracking technology.
作者
侯淋
杨顺华
黄时加
HOU Lin;YANG Shunhua;HUANG Shijia(China Aerodynamics Research and Development Center,Mianyang 621000,China;School of Electronics and Communication Engineering,Sun Yat-sen University,Guangzhou 511400,China)
出处
《电视技术》
2021年第4期21-24,共4页
Video Engineering
关键词
深度跟踪技术
判别模型
相关滤波
深度特征
deep tracking technology
discriminative model
correlation filter
deep feature