Human interaction prediction using deep temporal features

Qiuhong Ke; Mohammed Bennamoun; Senjian An; Farid Boussaid; Ferdous Sohel

doi:10.1007/978-3-319-48881-3_28

Human interaction prediction using deep temporal features

Qiuhong Ke, Mohammed Bennamoun, Senjian An, Farid Boussaid, Ferdous Sohel

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

40 Citations (Scopus)

Abstract

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.

Original language	English
Title of host publication	Computer Vision – ECCV 2016 Workshops, Proceedings
Editors	Gang Hua, Herve Jegou
Publisher	Springer-Verlag London Ltd.
Pages	403-414
Number of pages	12
ISBN (Print)	9783319488806
DOIs	https://doi.org/10.1007/978-3-319-48881-3_28
Publication status	Published - 2016
Externally published	Yes
Event	European Conference on Computer Vision 2016 - Amsterdam, Netherlands Duration: 11 Oct 2016 → 14 Oct 2016 Conference number: 14th http://www.eccv2016.org/ https://link.springer.com/book/10.1007/978-3-319-46448-0 (Proceedings)

Publication series

Name	Lecture Notes in Computer Science
Volume	9914
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	European Conference on Computer Vision 2016
Abbreviated title	ECCV 2016
Country/Territory	Netherlands
City	Amsterdam
Period	11/10/16 → 14/10/16
Internet address	http://www.eccv2016.org/ https://link.springer.com/book/10.1007/978-3-319-46448-0 (Proceedings)

Keywords

CNN
Interaction prediction
Temporal convolution

Access to Document

10.1007/978-3-319-48881-3_28

Cite this

@inproceedings{a26954d5240a4991acd83e13718125ca,

title = "Human interaction prediction using deep temporal features",

abstract = "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.",

keywords = "CNN, Interaction prediction, Temporal convolution",

author = "Qiuhong Ke and Mohammed Bennamoun and Senjian An and Farid Boussaid and Ferdous Sohel",

note = "Funding Information: This work was partially supported by Australian Research Council grants DP150100294, DP110103336, and DE120102960. Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; European Conference on Computer Vision 2016, ECCV 2016 ; Conference date: 11-10-2016 Through 14-10-2016",

year = "2016",

doi = "10.1007/978-3-319-48881-3\_28",

language = "English",

isbn = "9783319488806",

series = "Lecture Notes in Computer Science",

publisher = "Springer-Verlag London Ltd.",

pages = "403--414",

editor = "Gang Hua and Herve Jegou",

booktitle = "Computer Vision – ECCV 2016 Workshops, Proceedings",

address = "Germany",

url = "http://www.eccv2016.org/, https://link.springer.com/book/10.1007/978-3-319-46448-0",

}

Ke, Q, Bennamoun, M, An, S, Boussaid, F & Sohel, F 2016, Human interaction prediction using deep temporal features. in G Hua & H Jegou (eds), Computer Vision – ECCV 2016 Workshops, Proceedings. Lecture Notes in Computer Science, vol. 9914, Springer-Verlag London Ltd., pp. 403-414, European Conference on Computer Vision 2016, Amsterdam, Netherlands, 11/10/16. https://doi.org/10.1007/978-3-319-48881-3_28

Human interaction prediction using deep temporal features. / Ke, Qiuhong; Bennamoun, Mohammed; An, Senjian et al.
Computer Vision – ECCV 2016 Workshops, Proceedings. ed. / Gang Hua; Herve Jegou. Springer-Verlag London Ltd., 2016. p. 403-414 (Lecture Notes in Computer Science; Vol. 9914).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Human interaction prediction using deep temporal features

AU - Ke, Qiuhong

AU - Bennamoun, Mohammed

AU - An, Senjian

AU - Boussaid, Farid

AU - Sohel, Ferdous

N1 - Conference code: 14th

PY - 2016

Y1 - 2016

N2 - 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.

AB - 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.

KW - CNN

KW - Interaction prediction

KW - Temporal convolution

UR - https://www.scopus.com/pages/publications/84996644801

U2 - 10.1007/978-3-319-48881-3_28

DO - 10.1007/978-3-319-48881-3_28

M3 - Conference Paper

AN - SCOPUS:84996644801

SN - 9783319488806

T3 - Lecture Notes in Computer Science

SP - 403

EP - 414

BT - Computer Vision – ECCV 2016 Workshops, Proceedings

A2 - Hua, Gang

A2 - Jegou, Herve

PB - Springer-Verlag London Ltd.

T2 - European Conference on Computer Vision 2016

Y2 - 11 October 2016 through 14 October 2016

ER -

Human interaction prediction using deep temporal features

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this