Recognising human interaction from videos by a discriminative model
This study addresses the problem of recognising human interactions between two people. The main difficulties lie in the partial occlusion of body parts and the motion ambiguity in interactions. The authors observed that the interdependencies existing at both the action level and the body part level...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2014-08-01
|
Series: | IET Computer Vision |
Subjects: | |
Online Access: | https://doi.org/10.1049/iet-cvi.2013.0042 |
_version_ | 1797684635324383232 |
---|---|
author | Yu Kong Wei Liang Zhen Dong Yunde Jia |
author_facet | Yu Kong Wei Liang Zhen Dong Yunde Jia |
author_sort | Yu Kong |
collection | DOAJ |
description | This study addresses the problem of recognising human interactions between two people. The main difficulties lie in the partial occlusion of body parts and the motion ambiguity in interactions. The authors observed that the interdependencies existing at both the action level and the body part level can greatly help disambiguate similar individual movements and facilitate human interaction recognition. Accordingly, they proposed a novel discriminative method, which model the action of each person by a large‐scale global feature and local body part features, to capture such interdependencies for recognising interaction of two people. A variant of multi‐class Adaboost method is proposed to automatically discover class‐specific discriminative three‐dimensional body parts. The proposed approach is tested on the authors newly introduced BIT‐interaction dataset and the UT‐interaction dataset. The results show that their proposed model is quite effective in recognising human interactions. |
first_indexed | 2024-03-12T00:33:33Z |
format | Article |
id | doaj.art-ced90619e6c240e4a0880aac5fe51f7f |
institution | Directory Open Access Journal |
issn | 1751-9632 1751-9640 |
language | English |
last_indexed | 2024-03-12T00:33:33Z |
publishDate | 2014-08-01 |
publisher | Wiley |
record_format | Article |
series | IET Computer Vision |
spelling | doaj.art-ced90619e6c240e4a0880aac5fe51f7f2023-09-15T10:11:08ZengWileyIET Computer Vision1751-96321751-96402014-08-018427728610.1049/iet-cvi.2013.0042Recognising human interaction from videos by a discriminative modelYu Kong0Wei Liang1Zhen Dong2Yunde Jia3Beijing Laboratory of Intelligent Information TechnologySchool of Computer ScienceBeijing Institute of TechnologyBeijing100081People's Republic of ChinaBeijing Laboratory of Intelligent Information TechnologySchool of Computer ScienceBeijing Institute of TechnologyBeijing100081People's Republic of ChinaBeijing Laboratory of Intelligent Information TechnologySchool of Computer ScienceBeijing Institute of TechnologyBeijing100081People's Republic of ChinaBeijing Laboratory of Intelligent Information TechnologySchool of Computer ScienceBeijing Institute of TechnologyBeijing100081People's Republic of ChinaThis study addresses the problem of recognising human interactions between two people. The main difficulties lie in the partial occlusion of body parts and the motion ambiguity in interactions. The authors observed that the interdependencies existing at both the action level and the body part level can greatly help disambiguate similar individual movements and facilitate human interaction recognition. Accordingly, they proposed a novel discriminative method, which model the action of each person by a large‐scale global feature and local body part features, to capture such interdependencies for recognising interaction of two people. A variant of multi‐class Adaboost method is proposed to automatically discover class‐specific discriminative three‐dimensional body parts. The proposed approach is tested on the authors newly introduced BIT‐interaction dataset and the UT‐interaction dataset. The results show that their proposed model is quite effective in recognising human interactions.https://doi.org/10.1049/iet-cvi.2013.0042human interaction recognitiondiscriminative modelpartial occlusionmotion ambiguityaction levelbody part level |
spellingShingle | Yu Kong Wei Liang Zhen Dong Yunde Jia Recognising human interaction from videos by a discriminative model IET Computer Vision human interaction recognition discriminative model partial occlusion motion ambiguity action level body part level |
title | Recognising human interaction from videos by a discriminative model |
title_full | Recognising human interaction from videos by a discriminative model |
title_fullStr | Recognising human interaction from videos by a discriminative model |
title_full_unstemmed | Recognising human interaction from videos by a discriminative model |
title_short | Recognising human interaction from videos by a discriminative model |
title_sort | recognising human interaction from videos by a discriminative model |
topic | human interaction recognition discriminative model partial occlusion motion ambiguity action level body part level |
url | https://doi.org/10.1049/iet-cvi.2013.0042 |
work_keys_str_mv | AT yukong recognisinghumaninteractionfromvideosbyadiscriminativemodel AT weiliang recognisinghumaninteractionfromvideosbyadiscriminativemodel AT zhendong recognisinghumaninteractionfromvideosbyadiscriminativemodel AT yundejia recognisinghumaninteractionfromvideosbyadiscriminativemodel |