首页>成果

  • 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

[期刊论文]

Group Activity Recognition by Using Effective Multiple Modality Relation Representation With Temporal-Spatial Attention

分享
编辑 删除 报错

作者:

Xu, Dezhong (Xu, Dezhong.) | Fu, Heng (Fu, Heng.) | Wu, Lifang (Wu, Lifang.) (学者:毋立芳) | 展开

收录:

EI Scopus SCIE

摘要:

Group activity recognition has received a great deal of interest because of its broader applications in sports analysis, autonomous vehicles, CCTV surveillance systems and video summarization systems. Most existing methods typically use appearance features and they seldom consider underlying interaction information. In this work, a technology of novel group activity recognition is proposed based on multi-modal relation representation with temporal-spatial attention. First, we introduce an object relation module, which processes all objects in a scene simultaneously through an interaction between their appearance feature and geometry, thus allowing the modeling of their relations. Second, to extract effective motion features, an optical flow network is fine-tuned by using the action loss as the supervised signal. Then, we propose two types of inference models, opt-GRU and relation-GRU, which are used to encode the object relationship and motion representation effectively, and form the discriminative frame-level feature representation. Finally, an attention-based temporal aggregation layer is proposed to integrate frame-level features with different weights and form effective video-level representations. We have performed extensive experiments on two popular datasets, and both have achieved state-of-the-art performance. The datasets are the Volleyball dataset and the Collective Activity dataset, respectively.

关键词:

Activity recognition Visualization Optical losses Optical fiber networks Task analysis attention relation representation Optical imaging Group activity recognition motion representation Feature extraction

作者机构:

  • [ 1 ] [Xu, Dezhong]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China
  • [ 2 ] [Fu, Heng]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China
  • [ 3 ] [Wu, Lifang]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China
  • [ 4 ] [Jian, Meng]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China
  • [ 5 ] [Wang, Dong]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China
  • [ 6 ] [Liu, Xu]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China

通讯作者信息:

  • 毋立芳

    [Wu, Lifang]Beijing Univ Technol, Fac Informat Technol, Beijing 100124, Peoples R China

电子邮件地址:

查看成果更多字段

相关文章:

来源 :

IEEE ACCESS

ISSN: 2169-3536

年份: 2020

卷: 8

页码: 65689-65698

3 . 9 0 0

JCR@2022

被引次数:

WoS核心集被引频次: 11

SCOPUS被引频次: 21

近30日浏览量: 1

归属院系:

在线人数/总访问数:234/4515818
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司