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

Learning to transform time series with a few examples.

Ali Rahimi1, Ben Recht, Trevor Darrell

  • 1Intel, Seattle, WA 98105, USA. ali.rahimi@intel.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 19, 2007
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

Maternal and obstetric determinants of prematurity and term low birth weight in Afghanistan: a hospital-based case-control study.

Scientific reports·2026
Same author

Sports injuries among male futsal players in Herat, Afghanistan: a cross-sectional study.

BMC sports science, medicine & rehabilitation·2026
Same author

Determinants of access to anticancer medicines in South Asia: a multimethod study.

BMJ open·2026
Same author

Knowledge, Attitudes, and Practices Related to Diabetes Mellitus Among Hospital Outpatients and Their Companions in Herat, Afghanistan: A Cross-Sectional Study.

Public health challenges·2026
Same author

Awareness of Risk Factors, Warning Signs, Beliefs, and Practices Toward Gastric Cancer Screening among the General Population in Herat, Afghanistan: A Cross-Sectional Study.

Public health nursing (Boston, Mass.)·2026
Same author

The Prevalence of Internet Gaming Disorder and Its Associated Factors Among University Students in Herat City, Afghanistan.

Public health challenges·2026
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

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

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

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

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

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

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

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

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

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

Learning Shape Anchors for Holistic Indoor Scene Understanding.

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

This study introduces a new semi-supervised regression algorithm for time series transformation, learning from few examples to track targets efficiently. The method is faster and requires less data than existing approaches for various tracking applications.

Area of Science:

  • Machine Learning
  • Computer Vision
  • Robotics

Background:

  • Time series transformation is crucial for many applications, including target tracking.
  • Existing methods often require extensive labeled data or task-specific implementations.
  • Learning transformations directly from data can improve efficiency and reduce manual effort.

Purpose of the Study:

  • To develop a novel semi-supervised regression algorithm for learning time series transformations.
  • To apply this algorithm to various tracking problems, such as pose estimation.
  • To demonstrate its efficiency and reduced data requirements compared to existing methods.

Main Methods:

  • A semi-supervised regression algorithm is proposed that learns a memoryless transformation from example input-output mappings.

Related Experiment Videos

  • The algorithm finds a smooth function that fits training data and generates output time series consistent with assumed dynamics.
  • The method is related to nonlinear system identification and manifold learning, offering a fast, closed-form solution.

    • Main Results:

    • The algorithm successfully learns time series transformations for tracking tasks.
    • Demonstrated effectiveness in tracking RFID tags, rigid objects, deformable bodies, and articulated bodies.
    • Achieved superior performance with significantly fewer examples compared to fully-supervised or dynamics-agnostic semi-supervised methods.

    Conclusions:

    • The developed algorithm provides an efficient and data-frugal approach to learning time series transformations for tracking.
    • It offers a flexible alternative to task-specific implementations, applicable to diverse tracking scenarios.
    • The integration of output time series dynamics into the learning process is key to its effectiveness.