Trajectory-based modeling of human actions with motion reference points

Human action recognition in videos is a challenging problem with wide applications. State-of-the-art approaches often adopt the popular bag-of-features representation based on isolated local patches or temporal patch trajectories, where motion patterns like object relationships are mostly discarded....

Full description

Saved in:
Bibliographic Details
Main Authors: JIANG, Yu-Gang, DAI, Qi, XUE, Xiangyang, LIU, Wei, NGO, Chong-wah
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2012
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/6676
https://ink.library.smu.edu.sg/context/sis_research/article/7679/viewcontent/LNCS_7576___Computer_Vision___ECCV_2012.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-7679
record_format dspace
spelling sg-smu-ink.sis_research-76792023-08-21T01:19:17Z Trajectory-based modeling of human actions with motion reference points JIANG, Yu-Gang DAI, Qi XUE, Xiangyang LIU, Wei NGO, Chong-wah Human action recognition in videos is a challenging problem with wide applications. State-of-the-art approaches often adopt the popular bag-of-features representation based on isolated local patches or temporal patch trajectories, where motion patterns like object relationships are mostly discarded. This paper proposes a simple representation specifically aimed at the modeling of such motion relationships. We adopt global and local reference points to characterize motion information, so that the final representation can be robust to camera movement. Our approach operates on top of visual codewords derived from local patch trajectories, and therefore does not require accurate foreground-background separation, which is typically a necessary step to model object relationships. Through an extensive experimental evaluation, we show that the proposed representation offers very competitive performance on challenging benchmark datasets, and combining it with the bag-of-features representation leads to substantial improvement. On Hollywood2, Olympic Sports, and HMDB51 datasets, we obtain 59.5%, 80.6% and 40.7% respectively, which are the best reported results to date. 2012-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/6676 info:doi/10.1007/978-3-642-33715-4_31 https://ink.library.smu.edu.sg/context/sis_research/article/7679/viewcontent/LNCS_7576___Computer_Vision___ECCV_2012.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Action Recognition Human Action Recognition Histogram Intersection Dense Trajectory Video Stabilization Databases and Information Systems Graphics and Human Computer Interfaces
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Action Recognition
Human Action Recognition
Histogram Intersection
Dense Trajectory
Video Stabilization
Databases and Information Systems
Graphics and Human Computer Interfaces
spellingShingle Action Recognition
Human Action Recognition
Histogram Intersection
Dense Trajectory
Video Stabilization
Databases and Information Systems
Graphics and Human Computer Interfaces
JIANG, Yu-Gang
DAI, Qi
XUE, Xiangyang
LIU, Wei
NGO, Chong-wah
Trajectory-based modeling of human actions with motion reference points
description Human action recognition in videos is a challenging problem with wide applications. State-of-the-art approaches often adopt the popular bag-of-features representation based on isolated local patches or temporal patch trajectories, where motion patterns like object relationships are mostly discarded. This paper proposes a simple representation specifically aimed at the modeling of such motion relationships. We adopt global and local reference points to characterize motion information, so that the final representation can be robust to camera movement. Our approach operates on top of visual codewords derived from local patch trajectories, and therefore does not require accurate foreground-background separation, which is typically a necessary step to model object relationships. Through an extensive experimental evaluation, we show that the proposed representation offers very competitive performance on challenging benchmark datasets, and combining it with the bag-of-features representation leads to substantial improvement. On Hollywood2, Olympic Sports, and HMDB51 datasets, we obtain 59.5%, 80.6% and 40.7% respectively, which are the best reported results to date.
format text
author JIANG, Yu-Gang
DAI, Qi
XUE, Xiangyang
LIU, Wei
NGO, Chong-wah
author_facet JIANG, Yu-Gang
DAI, Qi
XUE, Xiangyang
LIU, Wei
NGO, Chong-wah
author_sort JIANG, Yu-Gang
title Trajectory-based modeling of human actions with motion reference points
title_short Trajectory-based modeling of human actions with motion reference points
title_full Trajectory-based modeling of human actions with motion reference points
title_fullStr Trajectory-based modeling of human actions with motion reference points
title_full_unstemmed Trajectory-based modeling of human actions with motion reference points
title_sort trajectory-based modeling of human actions with motion reference points
publisher Institutional Knowledge at Singapore Management University
publishDate 2012
url https://ink.library.smu.edu.sg/sis_research/6676
https://ink.library.smu.edu.sg/context/sis_research/article/7679/viewcontent/LNCS_7576___Computer_Vision___ECCV_2012.pdf
_version_ 1779156891396472832