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 Videos

Embedding motion in model-based stochastic tracking.

Jean-Marc Odobez1, Daniel Gatica-Perez, Sileye O Ba

  • 1IDIAP Research Institute, 1920 Martigny, Switzerland. odobez@idiap.ch

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|November 2, 2006
PubMed
Summary
This summary is machine-generated.

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

UveAI: clinic-ready scoring of retinal inflammation in uveitis on widefield fluorescein angiography using AI.

Scientific reports·2026
Same author

From Forest to Zoo: Great Ape Behavior Recognition with ChimpBehave.

International journal of computer vision·2025
Same author

Decoding community proximity discourse: A mixed-methods comparative analysis of online local and national newspapers in Romandy, Switzerland.

PloS one·2025
Same author

A systems approach towards remote health-monitoring in older adults: Introducing a zero-interaction digital exhaust.

NPJ digital medicine·2022
Same author

A Sensor-Driven Visit Detection System in Older Adults' Homes: Towards Digital Late-Life Depression Marker Extraction.

IEEE journal of biomedical and health informatics·2021
Same author

Towards an Engagement-Aware Attentive Artificial Listener for Multi-Party Interactions.

Frontiers in robotics and AI·2021

This study introduces a novel particle filtering approach for visual tracking. By incorporating motion, it improves sampling efficiency and prevents tracking failures caused by abrupt movements or visual distractors.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Robotics

Background:

  • Particle filtering is a popular visual tracking method.
  • Existing methods often assume temporal independence of observations and use transition priors, leading to inefficiencies and failures.
  • Abrupt motion changes and visual distractors pose significant challenges.

Purpose of the Study:

  • To address limitations in particle filtering for visual tracking.
  • To improve sampling efficiency and reduce tracking failures.
  • To enhance robustness against abrupt motion and visual distractors.

Main Methods:

  • Proposed a new particle filtering model incorporating motion.
  • Modeled conditional correlation of observations.
  • Introduced implicit/explicit motion measurements into the likelihood term.

Related Experiment Videos

  • Utilized explicit motion measurements to guide the sampling process.

    • Main Results:

    • The new model effectively handles abrupt motion changes.
    • It successfully filters out visual distractors.
    • Demonstrated superior tracking performance compared to the Condensation Algorithm in head tracking experiments with challenging dynamics.

    Conclusions:

    • The proposed motion-based particle filtering method offers improved visual tracking.
    • It enhances robustness and efficiency, particularly in dynamic scenarios.
    • This approach represents a significant advancement over traditional methods like the Condensation Algorithm.