This paper discusses the task of continuous human action recognition. By continuous, it refers to videos that contain multiple actions which are connected together. This task is important to applications like video surveillance and content based video retrieval. It aims to identify the action category and detect the start and end key frame of each action. It is a challenging task due to the frequent changes of human actions and the ambiguity of action boundaries. In this paper, a novel and efficient continuous action recognition framework is proposed. Our approach is based on the bag of words representation. A visual local pattern is regarded as a word and the action is modeled by the distribution of words. A generative translation and scale invariant probabilistic Latent Semantic Analysis model is presented. The continuous action recognition result is obtained frame by frame and updated from time to time. Experimental results show that this approach is effective and efficient to recognize both isolated actions and continuous actions.
%0 Journal Article
%1 GuoMiaoEtAl14mtaa
%A Guo, Ping
%A Miao, Zhenjiang
%A Shen, Yuan
%A Xu, Wanru
%A Zhang, Dianyong
%D 2014
%J Multimedia Tools and Applications
%K v1205 springer paper ai semantic processing image video action recognition analysis learn zzz.vitra
%N 3
%P 827-844
%R 10.1007/s11042-012-1084-2
%T Continuous Human Action Recognition in Real Time
%V 68
%X This paper discusses the task of continuous human action recognition. By continuous, it refers to videos that contain multiple actions which are connected together. This task is important to applications like video surveillance and content based video retrieval. It aims to identify the action category and detect the start and end key frame of each action. It is a challenging task due to the frequent changes of human actions and the ambiguity of action boundaries. In this paper, a novel and efficient continuous action recognition framework is proposed. Our approach is based on the bag of words representation. A visual local pattern is regarded as a word and the action is modeled by the distribution of words. A generative translation and scale invariant probabilistic Latent Semantic Analysis model is presented. The continuous action recognition result is obtained frame by frame and updated from time to time. Experimental results show that this approach is effective and efficient to recognize both isolated actions and continuous actions.
@article{GuoMiaoEtAl14mtaa,
abstract = {This paper discusses the task of continuous human action recognition. By continuous, it refers to videos that contain multiple actions which are connected together. This task is important to applications like video surveillance and content based video retrieval. It aims to identify the action category and detect the start and end key frame of each action. It is a challenging task due to the frequent changes of human actions and the ambiguity of action boundaries. In this paper, a novel and efficient continuous action recognition framework is proposed. Our approach is based on the bag of words representation. A visual local pattern is regarded as a word and the action is modeled by the distribution of words. A generative translation and scale invariant probabilistic Latent Semantic Analysis model is presented. The continuous action recognition result is obtained frame by frame and updated from time to time. Experimental results show that this approach is effective and efficient to recognize both isolated actions and continuous actions.},
added-at = {2014-02-07T17:21:39.000+0100},
author = {Guo, Ping and Miao, Zhenjiang and Shen, Yuan and Xu, Wanru and Zhang, Dianyong},
biburl = {https://www.bibsonomy.org/bibtex/2c913f297111d3ffff1b7ad6772304670/flint63},
doi = {10.1007/s11042-012-1084-2},
file = {SpringerLink:2014/GuoMiaoEtAl14mtaa.pdf:PDF},
groups = {public},
interhash = {a7dea042207c5a0bdf123b49b9ab6215},
intrahash = {c913f297111d3ffff1b7ad6772304670},
issn = {1380-7501},
journal = {Multimedia Tools and Applications},
keywords = {v1205 springer paper ai semantic processing image video action recognition analysis learn zzz.vitra},
month = {#feb#},
number = 3,
pages = {827-844},
timestamp = {2018-04-16T12:05:49.000+0200},
title = {Continuous Human Action Recognition in Real Time},
username = {flint63},
volume = 68,
year = 2014
}