Evaluation of vision-based human activity recognition in dense trajectory framework

Hirokatsu Kataoka, Yoshimitsu Aoki, Kenji Iwata, Yutaka Satoh

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

7 Citations (Scopus)


Activity recognition has been an active research topic in computer vision. Recently, the most successful approaches use dense trajectories that extract a large number of trajectories and features on the trajectories into a codeword. In this paper, we evaluate various features in the framework of dense trajectories on several types of datasets. We implement 13 features in total by including five different types of descriptor, namely motion-, shape-, texture- trajectory- and co-occurrence-based feature descriptors. The experimental results show a relationship between feature descriptors and performance rate at each dataset. Different scenes of traffic, surgery, daily living and sports are used to analyze the feature characteristics. Moreover, we test how much the performance rate of concatenated vectors depends on the type, top-ranked in experiment and all 13 feature descriptors on fine-grained datasets. Feature evaluation is beneficial not only in the activity recognition problem, but also in other domains in spatio-temporal recognition.

Original languageEnglish
Title of host publicationAdvances in Visual Computing - 11th International Symposium, ISVC 2015, Proceedings
EditorsMark Elendt, Richard Boyle, Eric Ragan, Bahram Parvin, Rogerio Feris, Tim McGraw, Ioannis Pavlidis, Regis Kopper, George Bebis, Darko Koracin, Zhao Ye, Gunther Weber
PublisherSpringer Verlag
Number of pages13
ISBN (Print)9783319278568
Publication statusPublished - 2015
Event11th International Symposium on Advances in Visual Computing, ISVC 2015 - Las Vegas, United States
Duration: 2015 Dec 142015 Dec 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other11th International Symposium on Advances in Visual Computing, ISVC 2015
Country/TerritoryUnited States
CityLas Vegas

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Evaluation of vision-based human activity recognition in dense trajectory framework'. Together they form a unique fingerprint.

Cite this