Related Experiment Video
Updated: May 21, 2026
08:19
Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion
Published on: January 15, 2016
Gait recognition across various walking speeds using higher order shape configuration based on a differential
Worapan Kusakunniran1, Qiang Wu, Jian Zhang
1School of Computer Science and Engineering, University of New South Wales (UNSW), Sydney, NSW 2052, Australia. worapan.kusakunniran@nicta.com.au
Summary
This study introduces a novel gait recognition method that accounts for walking speed variations. The differential composition model (DCM) effectively identifies individuals despite changes in speed, outperforming existing techniques.
Area of Science:
- Biometrics
- Human-Computer Interaction
- Computer Vision
Background:
- Gait is a reliable biometric for remote identification.
- Variations in walking speed significantly challenge gait recognition accuracy.
- Existing methods struggle to maintain performance across different speeds.
Purpose of the Study:
- To analyze the impact of walking speed variations on gait patterns.
- To develop a robust gait recognition method tolerant to speed changes.
- To improve the accuracy and reliability of gait identification systems.
Main Methods:
- Utilized Procrustes shape analysis for gait signature description and similarity measurement.
- Proposed a higher-order shape configuration for gait description, preserving discriminative information.
- Developed a differential composition model (DCM) to analyze speed effects on body parts and their discriminability.
Main Results:
- The proposed higher-order shape configuration effectively tolerates varying walking speeds.
- The DCM accurately differentiates speed-induced effects across body parts.
- Experimental results on standard gait databases confirm the method's efficiency for cross-speed recognition.
Conclusions:
- The novel gait recognition approach demonstrates superior performance in cross-speed scenarios.
- The differential composition model offers a robust solution for speed-invariant gait identification.
- This research advances the field of biometrics by addressing a critical challenge in gait recognition.
