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

Spatio-temporal context for robust multitarget tracking.

Hieu T Nguyen1, Qiang Ji, Arnold W M Smeulders

  • 1Department of Electrical, Computer, and Systems Engineering, Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY 12180, USA. nguyen@ecse.rpi.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|November 17, 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

Tissue-Specific Redistribution of Free Amino Acids in Mandarin Fish (<i>Siniperca chuatsi</i>) Under Acute Salinity, Alkalinity and Combined Saline-Alkaline Stress.

Life (Basel, Switzerland)·2026
Same author

Interpretable machine learning for prognostic prediction in young women with triple-negative invasive ductal breast cancer: a Boruta-SHAP integrated approach.

Gland surgery·2026
Same author

Respiratory chain gene mutations associated with global phylogenetic clustering of drug-resistant <i>Mycobacterium tuberculosis</i> revealed by whole-genome sequencing.

Frontiers in immunology·2026
Same author

An Image-based Global Climate Sentiment Index.

Scientific data·2026
Same author

High-Entropy Oxide Interlayers Enable Coordinated Regulation of Zn<sup>2+</sup> Interfacial Kinetics for Stable Zinc Metal Anodes.

ACS applied materials & interfaces·2026
Same author

Prognostic value of alkaline phosphatase-to-albumin ratio in critically ill patients with atrial fibrillation: A MIMIC-IV retrospective study.

Science progress·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Achieving Text-based Person Retrieval with Any Granularity.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a novel multitarget tracking method using spatial and temporal context. It enhances target identification accuracy, even with occlusions and similar targets, improving overall tracking robustness.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Maintaining target identity in multitarget tracking is challenging due to occlusions and target similarity.
  • Existing methods struggle with complex scenarios involving clutter and pose variations.

Purpose of the Study:

  • To develop an advanced multitarget tracking model that leverages contextual information for improved identity preservation.
  • To address limitations in current tracking systems concerning occlusions and subtle target differences.

Main Methods:

  • A novel multitarget tracking model incorporating spatial context (local background, nearby targets) and temporal context (historical appearances).
  • Target classification against spatial context, searching similar regions while avoiding nearby targets.

Related Experiment Videos

  • Integration of temporal context using Probabilistic Principal Component Analysis (PPCA) with an incremental online learning scheme.

    • Main Results:

    • The proposed method demonstrates robust tracking performance in scenarios with severe clutter.
    • Effective handling of occlusions and significant pose changes was observed.
    • Accurate online learning of PPCA parameters was achieved through a new incremental scheme.

    Conclusions:

    • Incorporating both spatial and temporal context significantly enhances multitarget tracking accuracy and robustness.
    • The developed model offers a promising solution for real-world tracking applications facing challenging conditions.
    • The PPCA-based approach with online learning provides an effective way to manage temporal target appearance information.