Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 26, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

Visual event recognition in videos by learning from Web data.

Lixin Duan1, Dong Xu, Ivor Wai-Hung Tsang

  • 1Nanyang Technological University, N4-02a-29, Nanyang Avenue, Singapore 639798. S080003@ntu.edu.sg

IEEE Transactions on Pattern Analysis and Machine Intelligence
|December 28, 2011
PubMed
Summary

This study introduces a visual event recognition framework using web videos to improve consumer video analysis. The Adaptive Multiple Kernel Learning (A-MKL) method effectively transfers knowledge, requiring fewer labeled consumer videos.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

MUFOLD-DB: a processed protein structure database for protein structure prediction and analysis.

BMC genomics·2015
Same author

Advances in translational bioinformatics facilitate revealing the landscape of complex disease mechanisms.

BMC bioinformatics·2015
Same author

The I-TASSER Suite: protein structure and function prediction.

Nature methods·2014
Same author

Genome-wide expression analysis of soybean NF-Y genes reveals potential function in development and drought response.

Molecular genetics and genomics : MGG·2014
Same author

Classification of lung cancer using ensemble-based feature selection and machine learning methods.

Molecular bioSystems·2014
Same author

Resveratrol possesses protective effects in a pristane-induced lupus mouse model.

PloS one·2014

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Consumer videos exhibit significant intra-class variations, posing challenges for event recognition.
  • Leveraging large, loosely labeled web video datasets is a promising approach for improving consumer video analysis.

Purpose of the Study:

  • To develop a visual event recognition framework for consumer videos using web data.
  • To address the challenge of domain shift between web and consumer video data.
  • To propose novel methods for video clip distance measurement and transfer learning.

Main Methods:

  • Aligned Space-Time Pyramid Matching (ASTPM) for measuring video clip dissimilarity.
  • Adaptive Multiple Kernel Learning (A-MKL) to fuse multi-level features and handle domain variations.

Related Experiment Videos

Last Updated: May 26, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

  • Training SVM classifiers on combined domain data and adapting them using prelearned classifiers.

    • Main Results:

    • The proposed framework effectively recognizes visual events in consumer videos.
    • A-MKL demonstrates superior performance by leveraging prelearned classifiers from all event classes.
    • The framework requires a minimal number of labeled consumer videos, showcasing efficient knowledge transfer.

    Conclusions:

    • The developed visual event recognition framework offers a robust solution for analyzing consumer videos.
    • Adaptive Multiple Kernel Learning is effective in bridging the domain gap between web and consumer videos.
    • The approach significantly reduces the need for extensive labeled consumer video data.