Feature Selection of Motion Capture Data in Gait Identification Challenge Problem

Świtoński, Adam; Josiński, Henryk; Michalczuk, Agnieszka; Pruszowski, Przemysław; Wojciechowski, Konrad

doi:10.1007/978-3-319-05458-2_55

Cited by 3 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To validate the proposed feature extraction, a motion database is collected. It contains data from 436 gaits of 30 different males between the ages of 20-40 (the database was also used in [2]). The gait route is a straight line of about five meters long.…”

Section: Classification Gait Databasementioning

confidence: 99%

“…Paper describes the new feature extraction for the classification of motion capture sequences. This classification task is challenging due to the data being complex, high dimensional and time-variant [1,2]. Motion data signals are decomposed into simpler representations in different scales using the second generation wavelet based multiresolution technique.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Wavelet features in motion data classification

Szczęsna

Świtoński

Słupik

et al. 2016

AIP Conference Proceedings

Self Cite

View full text Add to dashboard Cite

The paper deals with the problem of motion data classification based on result of multiresolution analysis implemented in form of quaternion lifting scheme. Scheme processes directly on time series of rotations coded in form of unit quaternion signal. In the work new features derived from wavelet energy and entropy are proposed. To validate the approach gait database containing data of 30 different humans is used. The obtained results are satisfactory. The classification has over than 91% accuracy.

show abstract

Section: Classification Gait Databasementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%