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

Matching shape sequences in video with applications in human movement analysis.

Ashok Veeraraghavan1, Amit K Roy-Chowdhury, Rama Chellappa

  • 1Center for Automation Research, #4417 A V Williams Building, University of Maryland at College Park, College Park, MD 20742, USA. vashok@umiacs.umd.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|December 17, 2005
PubMed
Summary

This study introduces novel methods for analyzing deforming shapes, enhancing gait-based human recognition. The approach utilizes both parametric and nonparametric models to accurately capture shape changes for improved person authentication.

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

Reply to the Letter to the Editor: "A perspective on AI implementation in medical imaging in LMICs: challenges, priorities, and strategies".

European radiology·2026
Same author

Deep Learning for the Detection of Corneal Perforation on Anterior-Segment Optical Coherence Tomography in Microbial Keratitis.

Bioengineering (Basel, Switzerland)·2026
Same author

Deep Learning for Detection of Corneal Perforation on Anterior Segment Optical Coherence Tomography in Microbial Keratitis.

medRxiv : the preprint server for health sciences·2026
Same author

Pose Guided Unsupervised Domain Adaptation for Human Body Part Segmentation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Accurate and fast event-based shape measurement of mixed reflectance scenes.

Nature communications·2026
Same author

Deep-learning endomicroscope with large field-of-view and depth-of-field for real-time in vivo imaging of epithelial cancer hallmarks.

Proceedings of the National Academy of Sciences of the United States of America·2026

Area of Science:

  • Computer Vision
  • Biometrics
  • Pattern Recognition

Background:

  • Comparing deforming shapes is crucial for various applications, including human recognition.
  • Existing methods often struggle with the complex, non-Euclidean nature of shape deformations.

Purpose of the Study:

  • To develop and evaluate novel methods for comparing sequences of deforming shapes.
  • To improve the accuracy of gait-based human recognition using shape deformation analysis.

Main Methods:

  • Utilized Kendall's definition of shape for feature extraction.
  • Proposed parametric models (autoregressive, autoregressive moving average) on the tangent space for shape analysis.
  • Adapted Dynamic Time-Warping (DTW) for non-Euclidean shape deformation spaces.

Related Experiment Videos

Main Results:

  • Parametric models effectively captured the nature of shape deformations in experiments.
  • The modified DTW algorithm demonstrated efficacy in gait-based human recognition.
  • Shape deformations of a person's silhouette proved to be a discriminating feature for authentication.

Conclusions:

  • The proposed approach offers a robust framework for comparing deforming shapes.
  • Shape and kinematics play significant roles in automated gait-based person authentication.
  • The study highlights the potential of non-Euclidean shape analysis in biometrics.