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

Improved gait recognition by gait dynamics normalization.

Zongyi Liu1, Sudeep Sarkar

  • 1Computer Science and Engineering Department, University of South Florida, 4202 E. Fowler Ave, ENB 118, Tampa, FL 33620, USA. zliu4@cse.usf.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 27, 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

Natural small molecule evodiamine alleviates liver fibrosis by targeting the ANXA2-p110α axis in hepatic stellate cells.

Phytomedicine : international journal of phytotherapy and phytopharmacology·2026
Same author

Optimal defocus for phase diversity in wavefront measurement of rectangular aperture system.

Optics express·2026
Same author

Land Finance and fiscal rules: An estimated DSGE model for Hong Kong.

PloS one·2026
Same author

Electrophilic (Hetero)Arene C-H Nitration by Ferric Nitrate Nonahydrate in Hexafluoroisopropanol.

The Journal of organic chemistry·2025
Same author

<i>Ipso</i>-Nitration of Boronic Esters Enabled by Ferric Nitrate Nonahydrate (Fe(NO<sub>3</sub>)<sub>3</sub>·9H<sub>2</sub>O) in HFIP.

Organic letters·2025
Same author

Automated pressure ulcer dimension measurements using a depth camera.

Journal of wound care·2025
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 enhances gait recognition by normalizing walking dynamics and focusing on gait shape. Improved accuracy was demonstrated on multiple datasets, even with varying conditions.

Area of Science:

  • Biometrics
  • Computer Vision
  • Pattern Recognition

Background:

  • Gait biometrics leverage unique walking patterns for identification.
  • Existing methods often struggle with variations in gait dynamics and appearance.

Purpose of the Study:

  • To improve gait recognition accuracy by separating and normalizing gait dynamics.
  • To focus on gait shape information for more robust identification.

Main Methods:

  • Utilized a population Hidden Markov Model (pHMM) to normalize gait dynamics.
  • Employed Viterbi decoding to obtain a dynamics-normalized gait cycle.
  • Computed distances between gait silhouettes in a Linear Discriminant Analysis (LDA) space, invariant to scale and imaging variations.

Main Results:

Related Experiment Videos

  • Significantly improved gait recognition performance on the HumanID Gait Challenge dataset, particularly under challenging conditions like surface changes and carrying objects.
  • Demonstrated enhanced performance on the UMD and CMU Mobo datasets, including matching across different walking speeds without specific training.

Conclusions:

  • Normalizing gait dynamics and emphasizing shape information is a superior approach for gait recognition.
  • The proposed method offers robustness against variations in walking conditions and imaging, proving effective across diverse datasets.