IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2016-03-24 10:00
Temporal Spotting of Human Actions in Video Based on Voting Framework with a Hierarchical Action Model
Keita Hara, Kazuaki Nakamura, Noboru Babaguchi (Osaka Univ.) BioX2015-42 PRMU2015-165
Abstract (in Japanese) (See Japanese page) 
(in English) This paper focuses on the task of temporal action spotting: temporal segmentation and classification of human actions in video. Existing methods for temporal action spotting can be divided into two types: voting-based and DTW-based. The former is robust to the motions that are unrelated to targets actions, whereas its performance is sensitive to the speed of actions. In contrast, the latter is robust to different action speed but sensitive to the motions unrelated to target actions. In this paper, we propose a method that can deal with both problems by combining voting and DTW. To this end, we introduce a hierarchical action model consisting of frame-wise visual features, target actions, and the intermediate layer called ``motion primitives'' whose length is different.
Our proposed method first converts a sequence of visual features to that of motion primitives using DTW, and then applies a voting-based method, regarding each motion primitive as a unit of time. Experimental results showed that the performance of the proposed method is less degraded by the above two problems compared with existing methods.
Keyword (in Japanese) (See Japanese page) 
(in English) temporal action spotting / action recognition / temporal action segmentation / motions unrelated to target actions / different speed of action / motion primitives / /  
Reference Info. IEICE Tech. Rep., vol. 115, no. 517, PRMU2015-165, pp. 7-12, March 2016.
Paper # PRMU2015-165 
Date of Issue 2016-03-17 (BioX, PRMU) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF BioX2015-42 PRMU2015-165

Conference Information
Committee PRMU BioX  
Conference Date 2016-03-24 - 2016-03-25 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To PRMU 
Conference Code 2016-03-PRMU-BioX 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Temporal Spotting of Human Actions in Video Based on Voting Framework with a Hierarchical Action Model 
Sub Title (in English)  
Keyword(1) temporal action spotting  
Keyword(2) action recognition  
Keyword(3) temporal action segmentation  
Keyword(4) motions unrelated to target actions  
Keyword(5) different speed of action  
Keyword(6) motion primitives  
Keyword(7)  
Keyword(8)  
1st Author's Name Keita Hara  
1st Author's Affiliation Osaka University (Osaka Univ.)
2nd Author's Name Kazuaki Nakamura  
2nd Author's Affiliation Osaka University (Osaka Univ.)
3rd Author's Name Noboru Babaguchi  
3rd Author's Affiliation Osaka University (Osaka Univ.)
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2016-03-24 10:00:00 
Presentation Time 30 minutes 
Registration for PRMU 
Paper # BioX2015-42, PRMU2015-165 
Volume (vol) vol.115 
Number (no) no.516(BioX), no.517(PRMU) 
Page pp.7-12 
#Pages
Date of Issue 2016-03-17 (BioX, PRMU) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan