Introduction Until recently, 3D tracking of human swimming motion capture (mo-cap) from an image has been conducted using interactive tracking (IT). New technology within automatic tracking (AT) in water uses computer algorithms to identify and track small reflective markers attached to the swimmer. The purpose of this study was to show that 3D AT can be used for measuring kinematic variables in human swimming such as the intra-cyclic velocity variations (dv) within a stroke cycle in breaststroke. Methods: Four swimmers (one male, world championship medalist and three females, one olympic medalist, one national medalist and one national finalist) performed one trial of 20m normal breaststroke at the speeds of 60, 70, 80, 90 and 100% of maximal effort. During each trial, the dv of the swimmers body were recorded at the left and right pelvic and left and right trochanter major (glued to the swimmers suit) with ten underwater mocap cameras (Oqus Underwater, Qualisys, Gothenburg, Sweden). The cameras recorded spherical markers (diameter of 19 mm) attached to the swimmers using cyan LED light. A 2.5(z)x1.5(y)x10(x) volume was calibrated using a moving wand method. For statistical analysis trochanter major was selected due to a tracking recording of 100%. The second and third stroke cycle was selected to avoid excess speed from the push-off. The stroke cycle started when the heels were fully pulled up, flexed and ready to kick backward. dv was calculated as the difference between the highest and lowest velocity peaks within the stroke cycle, relative to the average velocity. Results: The highest and lowest velocity was on average recorded at 72.05% and 93,83% into the stroke cycle for the world class swimmers (WC) and at 63.03% and 96.36% for the national elite swimmers (NE). A t-test showed a significant difference among the groups for the highest dv p<0.001 and no differences for the lowest dv (p=0.07). R values of .959 for the WC and .859 for the NE showed strong correlations between average cycle velocity and the dv fluctuations. Discussion: We found that the highest velocity during the breaststroke cycle corresponded to approximately the middle of the insweep phase. Subjects had no movement constraints, but had significant added drag from wearing 36 reflective markers (Kjendlie et al., 2012) and also wearing sEMG equipment described by (Olstad et al., 2011). The study successfully showed that 3D analysis can be performed in water using AT. Data analysis with AT is time effective, and open up new possibilities for conducting wide-scale studies investigating different kinematic variables in swimming.
© Copyright 2012 17th Annual Congress of the European College of Sport Science (ECSS), Bruges, 4. -7. July 2012. Published by Vrije Universiteit Brussel. All rights reserved.