Action Recognition for Videos by Long-Term Point Trajectory Analysis with Background Removal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Recently, dense trajectories were shown to be an efficient video motion representation for action recognition and achieved state-of-the-art results on a variety of video datasets. This paper improves their performance by taking into account camera motion. To estimate camera motion, the authors use long-term point trajectory analysis to cluster image points and propose an algorithm to find possible background cluster from these clusters according to background nature in a video. Considering the original clusters could not segment the foreground and background very well. The authors optimize the background cluster, and use the cluster to rectify the trajectory. Experimental results on three challenging action datasets (i.e., Hollywood2, Olympic Sports and UCF50) show that the rectified trajectories significantly outperform original dense trajectories.

Original languageEnglish
Title of host publicationProceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
EditorsGiuseppe De Pietro, Albert Dipanda, Richard Chbeir, Luigi Gallo, Kokou Yetongnon
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages23-30
Number of pages8
ISBN (Electronic)9781509056989
DOIs
Publication statusPublished - Apr 21 2017
Event12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016 - Naples, Italy
Duration: Nov 28 2016Dec 1 2016

Publication series

NameProceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016

Other

Other12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
Country/TerritoryItaly
CityNaples
Period11/28/1612/1/16

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Radiology Nuclear Medicine and imaging
  • Computer Networks and Communications
  • Signal Processing

Fingerprint

Dive into the research topics of 'Action Recognition for Videos by Long-Term Point Trajectory Analysis with Background Removal'. Together they form a unique fingerprint.

Cite this