Quantitative underwater 3D motion analysis using submerged video cameras: accuracy analysis and trajectory reconstruction

Abstract

In this study we aim at investigating the applicability of underwater 3D motion capture based on submerged video cameras in terms of 3D accuracy analysis and trajectory reconstruction. Static points with classical direct linear transform (DLT) solution, a moving wand with bundle adjustment and a moving 2D plate with Zhang's method were considered for camera calibration. As an example of the final application, we reconstructed the hand motion trajectories in different swimming styles and qualitatively compared this with Maglischo's model. Four highly trained male swimmers performed butterfly, breaststroke and freestyle tasks. The middle fingertip trajectories of both hands in the underwater phase were considered. The accuracy (mean absolute error) of the two calibration approaches (wand: 0.96 mm - 2D plate: 0.73 mm) was comparable to out of water results and highly superior to the classical DLT results (9.74 mm). Among all the swimmers, the hands' trajectories of the expert swimmer in the style were almost symmetric and in good agreement with Maglischo's model. The kinematic results highlight symmetry or asymmetry between the two hand sides, intra- and inter-subject variability in terms of the motion patterns and agreement or disagreement with the model. The two outcomes, calibration results and trajectory reconstruction, both move towards the quantitative 3D underwater motion analysis.

Optimising filtering parameters for a 3D motion analysis system

“…Ceccon et al (2013) found a coefficient of variation of 7.58% for a 10 cm reference wand moving through the calibrated, underwater volume. Silvatti et al (2013) compared different calibration procedures and found a coefficient of variation as low as 0.24% for a 3 m wand, which is much larger than the rigid body used in our study to calculate dynamic precision. In contrast to the aforementioned studies in which a wand was moved through the calibrated volume, we studied the precision during the underwater phase of the stroke.…”

“…The movement of the body parts through the water generates extra bubbles, which most likely increases reconstruction errors due to additional refraction. In addition, different calibration procedures (Silvatti et al, 2013), camera positions and camera resolutions could have caused differences in precision between studies. As pointed out by Figueiredo et al (2011), the precision obtained in motion analysis systems for swimming is influenced by image distortion and by errors related to digitisation and 3D reconstruction (Payton and Bartlett, 1995;Kwon and Casebolt, 2006).…”

“…Moreover, experimental set-ups for underwater 3D reconstruction using video cameras (e.g. Ceccon et al, 2013;Silvatti et al, 2013) were found to be less precise compared to values obtained above water (Ehara et al, 1995). Therefore, underwater environments represent a context where optimising precision is of particular concern.…”

The role of the entry-and-stretch phase at the different paces of race in front crawl swimming

“…Indeed, the uncertainty of measurement is 0.033 s, which is the duration of one frame. Furthermore, the hypothesis of a symmetric arm coordination, made here, is not completely exact, because an asymmetry linked to the swimmer's technique (Silvatti et al, 2013;Tourny-Chollet, Seifert, & Chollet, 2009) or to the breathing (Seifert, 2010;Seifert, Chollet, & Allard, 2005) can exist.…”

Application of Video-Based Methods for Competitive Swimming Analysis: A Systematic Review

“…Others have provided an extensive examination of the technical aspects of underwater videography, with an emphasis on calibration and reconstruction procedures. [21][22][23] No review has been published specifically assessing the processes by which video is captured in applied swimming settings. This may result in uncertainty amongst coaches and practioners regarding the most appropriate methodologies to be adopted and the value of video in swimming.…”