RESUMEN
The purpose of this study was to compare the two main types of marker sets for human body representation based on rigid clusters of markers and skin surface markers for measuring kinematics during walking and running. Velocity, body segment, and joint angle were considered in the comparison of both protocols. Six male athletes were studied during treadmill gait at 1.4 and 5.5 m/s and recorded with 8 high speed video cameras. The subjects used simultaneously both protocols in the same walking and running cycles, in order to compare the variability in the determination of the joint centers' positions and the joint angles calculated from each protocol. The three-way ANOVA results showed that the variability of the inter-markers distance in the skin surface protocol was higher than that in the rigid clusters of markers, as reported in the literature. However, no statistical differences between the protocols were found in the variability of the determination of the joint centers' positions. Therefore no advantage was verified to rigid cluster protocols even for the upper body segments. Another conclusion is that increases in velocity produced increases in variability of the joint centers' distances and increases in the maximum differences between the joint angles.