Introduction The use of whole body 3D human geometric representations in 3D body scanning applications for commercial and scientific purposes (see Robinette and Daanen, 2006) as well as to investigate markerless motion tracking in scientific applications (Corazzaa, Mündermanna, and Andriacchi, 2007) has received much interest recently. However, the present approaches may have potential limitations (see Esteban and Schmitt, 2004). No studies have investigated both the accuracy and reliability of 3D reconstructed geometric models. Quah, Gagalowicz, and Seah (2007) described an approach in 3D human geometric model reconstruction derived from generic models where the common challenges associated with holes filling are mitigated. This approach is potentially cost effective and easy to set up but it has not been compared with other commonly used methods in terms of precision and accuracy. Methods Two participants of contrasting sizes volunteered for this study which was approved by the institution`s ethics committee. An active laser scanner (AS); a simple synchronized 4 fire-wire cameras cum desktop setup (FWS) and an active marker motion analysis system (AM) was used to capture the data. The participants were scanned in the normal anatomical standing position using the FWS and AS systems. They stood in the same position with markers placed on the body for the AM system to capture the markers` spatial coordinates. 3D geometric models were reconstructed by the AS system (using its bundled software) as well as the FWS system (using the method described by Quah et al, 2007). Various anthropometrical linear parameters were computed like wrist width, shoulder width and ankle heights. Results There was no significant intra-participant differences (p>0.05) in the parameters measured. Paired ttests showed no significant difference in the mean values between the methods (p>0.05) and the coefficient of variance was generally at below the 2% level. Discussion / Conclusion s The accuracy and precision of the FWS approach suggest that it is potentially feasible to use for 3D human geometric representations. This has significant impact and implications for future work in areas where interest in markerless motion capture applications is on the rise. Future studies involving more subjects with varying sizes and additional measured body parameters involving, say, circumferences, can be done to further understand the validity of the FWC approach.
© Copyright 2008 2008 International Convention on Science, Education and Medicine in Sport: Proceedings, Vol. II. Veröffentlicht von People´s Sports Publishing House. Alle Rechte vorbehalten.