摘要
目标跟踪是计算机视觉领域核心问题之一。近十年来目标跟踪技术取得较快发展,新理论、新方法、新工具层出不穷。基于中心距误差、成功率及最好结果三个度量标准,将当前视频跟踪中常用的八种目标跟踪算法应用于养殖场动物(例如奶牛)跟踪,并给出影响跟踪性能的详细分析。
Target tracking is one of the core problems in computer vision. In recent years, target tracking technology achieves rapid development and there have been a lot of new theories, new methods and new tools. Based on the metrics of success rate, center location error and the best result, analyzes the animal （like dairy cattle） motion using the state-of-the-art tracking methods, and gives some detailed discusses which affect the object tracking performance.
出处
《现代计算机》
2014年第12期3-7,共5页
Modern Computer
关键词
目标跟踪
分类器
机器学习
Target Tracking
Classification
Machine Learning
