摘要
针对人工智能系列"挑战性课程"建设在本科教育中的重要作用,结合计算机研究领域中应用广泛的"视频多目标跟踪"课题,以及前沿的"神经网络"技术,实现了算法性能领先的"视频多目标跟踪系统"的综合实验设计。通过实验结果的分析发现,基于递归神经网络和长短期记忆模型的多目标跟踪系统,能够有效地完成视频的多目标跟踪。
In view of the important role of the construction of artificial intelligence series of"Challenging"courses in undergraduate education,combined with the project of the"Video multi-target tracking"widely used in the field of computer research and as well as the cutting-edge"Neural network"technology,the comprehensive experimental design of the"Video multi-target tracking system"with leading algorithm performance is realized.Through the analysis of the experimental results,it is found that the multi-target tracking system based on recurrent neural network and short-term memory model can effectively complete the video multi-target tracking.
作者
陈娟
杨倩
文泉
刘歆浏
刘议聪
CHEN Juan;YANG Qian;WEN Quan;LIU Xinliu;LIU Yicong(School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China;Southwest Automation Research Institute,Mianyang 621000,China)
出处
《实验技术与管理》
CAS
北大核心
2020年第1期155-158,共4页
Experimental Technology and Management
基金
教育部“华为”产学研项目(201802001046).
关键词
多目标跟踪
递归神经网络
长短期记忆模型
数据关联
挑战性课程
multi-target tracking
recurrent neural network
long-short term memory model
data association
challenging courses
