The application of principal component analysis to quantify technique in sports

Analyzing an athlete's “technique,” sport scientists often focus on preselected variables that quantify important aspects of movement. In contrast, coaches and practitioners typically describe movements in terms of basic postures and movement components using subjective and qualitative features. A challenge for sport scientists is finding an appropriate quantitative methodology that incorporates the holistic perspective of human observers. Using alpine ski racing as an example, this study explores principal component analysis (PCA) as a mathematical method to decompose a complex movement pattern into its main movement components. Ski racing movements were recorded by determining the three‐dimensional coordinates of 26 points on each skier which were subsequently interpreted as a 78‐dimensional posture vector at each time point. PCA was then used to determine the mean posture and principal movements (PMk) carried out by the athletes. The first four PMk contained 95.5 ± 0.5% of the variance in the posture vectors which quantified changes in body inclination, vertical or fore‐aft movement of the trunk, and distance between skis. In summary, calculating PMk offered a data‐driven, quantitative, and objective method of analyzing human movement that is similar to how human observers such as coaches or ski instructors would describe the movement.

[1]  N. Troje,et al.  The Inversion Effect in Biological Motion Perception: Evidence for a “Life Detector”? , 2006, Current Biology.

[2]  A. Mizuno,et al.  A change of the leading player in flow Visualization technique , 2006, J. Vis..

[3]  P Scheiber,et al.  Relationships between biomechanics and physiology in older, recreational alpine skiers , 2012, Scandinavian journal of medicine & science in sports.

[4]  Erich Müller,et al.  Biomechanical aspects of new techniques in alpine skiing and ski-jumping , 2003, Journal of sports sciences.

[5]  N. Troje,et al.  Person identification from biological motion: Effects of structural and kinematic cues , 2005, Perception & psychophysics.

[6]  A. Lees Technique analysis in sports: a critical review , 2002, Journal of sports sciences.

[7]  Nikolaus F. Troje,et al.  3D Periodic Human Motion Reconstruction from 2D Motion Sequences , 2007, Neural Computation.

[8]  Renato Rodano,et al.  Application of functional principal component analysis in race walking: An emerging methodology , 2009, Sports biomechanics.

[9]  Vladimir M. Zatsiorsky,et al.  Kinetics of Human Motion , 2002 .

[10]  Mike D Hughes,et al.  The use of performance indicators in performance analysis , 2002, Journal of sports sciences.

[11]  P. Leva Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. , 1996 .

[12]  Cord Westhoff,et al.  Kinematic cues for person identification from biological motion , 2007, Perception & psychophysics.

[13]  Ruhr-Universitát Bochum Person Identification from Biological Motion , 2005 .

[14]  Arend L. Schwab,et al.  Rider motion identification during normal bicycling by means of principal component analysis , 2011 .

[15]  W. Rodgers,et al.  New Ideas for High Performance Coaches: A Case Study of Knowledge Transfer in Sport Science , 2008 .

[16]  S. Loland The Mechanics and Meaning of Alpine Skiing: Methodological and Epistemological Notes on the Study of Sport Technique , 1992 .

[17]  Yen-Lin Chen,et al.  3D Reconstruction of Human Motion and Skeleton from Uncalibrated Monocular Video , 2009, ACCV.

[18]  Kazuyoshi Takayama,et al.  Holographic interferometric visualization of the Richtmyer-Meshkov instability induced by cylindrical shock waves , 1999 .

[19]  W. Rodgers,et al.  Knowledge Transfer: How do High Performance Coaches Access the Knowledge of Sport Scientists? , 2008 .

[20]  Adri Hartveld MSc Mcsp Qualitative Analysis of Human Movement , 1998 .

[21]  Dorita H. F. Chang,et al.  Acceleration carries the local inversion effect in biological motion perception. , 2009, Journal of vision.

[22]  Katsu Yamane,et al.  Characterization of motor skill based on musculoskeletal model , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[23]  Peter Federolf,et al.  Finite element simulation of the ski–snow interaction of an alpine ski in a carved turn , 2010 .

[24]  H. M. Karara,et al.  Direct Linear Transformation from Comparator Coordinates into Object Space Coordinates in Close-Range Photogrammetry , 2015 .

[25]  Dorita H. F. Chang,et al.  Characterizing global and local mechanisms in biological motion perception. , 2009, Journal of vision.

[26]  P Scheiber,et al.  Impact of skier actions on the gliding times in alpine skiing , 2008, Scandinavian journal of medicine & science in sports.

[27]  Andreas Daffertshofer,et al.  PCA in studying coordination and variability: a tutorial. , 2004, Clinical biomechanics.

[28]  N. Troje Decomposing biological motion: a framework for analysis and synthesis of human gait patterns. , 2002, Journal of vision.

[29]  N. Troje,et al.  The facing bias in biological motion perception: Effects of stimulus gender and observer sex , 2010, Attention, perception & psychophysics.

[30]  David J. Fleet,et al.  Human attributes from 3D pose tracking , 2010, Comput. Vis. Image Underst..

[31]  E. Müller,et al.  Instructor-Paced vs. Self-Paced Skiing Modes in Older Recreational Alpine Skiers , 2011, Journal of strength and conditioning research.

[32]  N. Troje,et al.  Differences in Gait Across the Menstrual Cycle and Their Attractiveness to Men , 2008, Archives of sexual behavior.