基于深度学习的行为检测方法综述
作者:
作者单位:

作者简介:

通讯作者:

基金项目:

国家自然科学基金(61571071, 61906025);重庆市科委自然科学基金(cstc2018jcyjAX0227)


Deep learning based action detection: a survey
Author:
Affiliation:

Fund Project:

The National Natural Science Foundation of China(61571071, 61906025); The Natural Science Foundation Project of Chongqing Science and Technology(cstc2018jcyjAX0227)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    行为检测在自动驾驶、视频监控等领域的广阔应用前景使其成为了视频分析的研究热点。近年来,基于深度学习的方法在行为检测领域取得了巨大的进展,引起了国内外研究者的关注,对这些方法进行了全面的梳理和总结,介绍了行为检测任务的详细定义和面临的主要挑战;从时序行为检测和时空行为检测2个方面对相关文献做了细致地分类,综合分析了每一类别中不同研究方法的思路和优缺点,并阐述了基于弱监督学习、图卷积神经网络、注意力机制等新兴研究话题的相关方法;介绍了行为检测领域常用的数据集以及性能评估指标,在这些数据集上比较了几种典型方法的性能;总结了当前行为检测方法需要解决的问题以及进一步发展的研究方向。

    Abstract:

    Action detection becomes a research hotspot in video analysis due to its broad application prospects in autonomous driving, video surveillance, etc. In recent years, methods based on deep learning have made great progress in the field of action detection, and have attracted the attention of researchers at home and abroad. This paper summarizes these methods comprehensively. Firstly, the definition and challenges of the action detection task are introduced. Then, we classify relevant literature carefully from two aspects:temporal action detection and spatio-temporal action detection. The ideas, advantages, and disadvantages of different methods in each category are comprehensively analyzed. Additionally, we introduce some methods based on hot technologies such as weakly supervised learning, graph convolutional network,attention mechanism.Some of the most commonly used datasets and metrics are listed and the performances of the typical methods are compared on these datasets. Finally, we summarize the problems to be solved in the future and some directions worthy of attention for action detection community.

    参考文献
    相似文献
    引证文献
引用本文

高陈强,陈旭.基于深度学习的行为检测方法综述[J].重庆邮电大学学报(自然科学版),2020,32(6):991-102.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2019-12-02
  • 最后修改日期:2020-03-11
  • 录用日期:
  • 在线发布日期: 2020-12-23

微信公众号二维码

手机版网站二维码