Presentation 2015-02-20
Human Action Retrieval Based on Temporal Matching
Mengxi LIN, Nakamasa INOUE, Koichi SHINODA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper focuses on human action retrieval, which aims to retrieve video segments from a video database based on an action of interest specified on-the-fly by a user. In this work, we focus on capturing temporal information of actions and propose to utilize Dynamic Time Warping (DTW) to measure the temporal distortion and difference between a pair of actions. Temporal motion saliency of a query video is introduced to re-rank the retrieval results. We evaluate our method in the Breakfast dataset and show that our method is more effective than the baselines, which do not consider any temporal orders of actions.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) action retrieval / action localization / temporal matching / DTW
Paper # PRMU2014-139,CNR2014-54
Date of Issue

Conference Information
Committee CNR
Conference Date 2015/2/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Cloud Network Robotics (CNR)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Human Action Retrieval Based on Temporal Matching
Sub Title (in English)
Keyword(1) action retrieval
Keyword(2) action localization
Keyword(3) temporal matching
Keyword(4) DTW
1st Author's Name Mengxi LIN
1st Author's Affiliation Department of Computer Science, Tokyo Institute of Technology()
2nd Author's Name Nakamasa INOUE
2nd Author's Affiliation Department of Computer Science, Tokyo Institute of Technology
3rd Author's Name Koichi SHINODA
3rd Author's Affiliation Department of Computer Science, Tokyo Institute of Technology
Date 2015-02-20
Paper # PRMU2014-139,CNR2014-54
Volume (vol) vol.114
Number (no) 455
Page pp.pp.-
#Pages 6
Date of Issue