Title :
Gait analysis and identification
Author :
Jie Hong ; Jinsheng Kang ; Price, M.E.
Author_Institution :
Sch. of Eng. & Design, Brunel Univ., Uxbridge, UK
Abstract :
An efficient framework were proposed for identifying individuals from gait via feature-based method based on 3D motion capture data. Three different extraction methods were applied to achieve gait signatures. The average identification rate was over 93% with best result close to 100% in a 35 subject database. In additional, gait attractiveness was analyzed via Principle component analysis and linear regression method. A systematic relationship was found between the motions of individual markers and the attractiveness ratings. In a linear equation, ln(PCA1) and ln(PCA2) predicted ln(attract_value) with reasonable accuracy.
Keywords :
feature extraction; gait analysis; image motion analysis; principal component analysis; regression analysis; 3D motion capture data; attractiveness ratings; feature-based method; gait analysis; gait identification; gait signatures; individual identification framework; linear equation; linear regression method; principal component analysis; systematic relationship; Equations; Feature extraction; Humans; Legged locomotion; Linear regression; Pattern recognition; Principal component analysis; Gait identify; Linear regression; Principle component analysis; gait signature;
Conference_Titel :
Automation and Computing (ICAC), 2012 18th International Conference on
Conference_Location :
Loughborough
Print_ISBN :
978-1-4673-1722-1
