Human interaction prediction using deep temporal features

© Springer International Publishing Switzerland 2016. Interaction prediction has a wide range of applications such as robot controlling and prevention of dangerous events. In this paper, we introduce a new method to capture deep temporal information in videos for human interaction prediction. We pro...

Full description

Bibliographic Details
Main Authors: Ke, Q., Bennamoun, M., An, Senjian, Boussaid, F., Sohel, F.
Format: Conference Paper
Published: 2016
Online Access:http://hdl.handle.net/20.500.11937/69567
_version_ 1848762075278475264
author Ke, Q.
Bennamoun, M.
An, Senjian
Boussaid, F.
Sohel, F.
author_facet Ke, Q.
Bennamoun, M.
An, Senjian
Boussaid, F.
Sohel, F.
author_sort Ke, Q.
building Curtin Institutional Repository
collection Online Access
description © Springer International Publishing Switzerland 2016. Interaction prediction has a wide range of applications such as robot controlling and prevention of dangerous events. In this paper, we introduce a new method to capture deep temporal information in videos for human interaction prediction. We propose to use flow coding images to represent the low-level motion information in videos and extract deep temporal features using a deep convolutional neural network architecture. We tested our method on the UT-Interaction dataset and the challenging TV human interaction dataset, and demonstrated the advantages of the proposed deep temporal features based on flow coding images. The proposed method, though using only the temporal information, outperforms the state of the art methods for human interaction prediction.
first_indexed 2025-11-14T10:41:48Z
format Conference Paper
id curtin-20.500.11937-69567
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T10:41:48Z
publishDate 2016
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-695672018-08-08T04:57:02Z Human interaction prediction using deep temporal features Ke, Q. Bennamoun, M. An, Senjian Boussaid, F. Sohel, F. © Springer International Publishing Switzerland 2016. Interaction prediction has a wide range of applications such as robot controlling and prevention of dangerous events. In this paper, we introduce a new method to capture deep temporal information in videos for human interaction prediction. We propose to use flow coding images to represent the low-level motion information in videos and extract deep temporal features using a deep convolutional neural network architecture. We tested our method on the UT-Interaction dataset and the challenging TV human interaction dataset, and demonstrated the advantages of the proposed deep temporal features based on flow coding images. The proposed method, though using only the temporal information, outperforms the state of the art methods for human interaction prediction. 2016 Conference Paper http://hdl.handle.net/20.500.11937/69567 10.1007/978-3-319-48881-3_28 restricted
spellingShingle Ke, Q.
Bennamoun, M.
An, Senjian
Boussaid, F.
Sohel, F.
Human interaction prediction using deep temporal features
title Human interaction prediction using deep temporal features
title_full Human interaction prediction using deep temporal features
title_fullStr Human interaction prediction using deep temporal features
title_full_unstemmed Human interaction prediction using deep temporal features
title_short Human interaction prediction using deep temporal features
title_sort human interaction prediction using deep temporal features
url http://hdl.handle.net/20.500.11937/69567