Skip to main content

Table 2 Mean (± 1SD) RMSE (% BW) achieved by the three models from estimation of all three components of GRF

From: Performance of machine learning models in estimation of ground reaction forces during balance exergaming

  3DMoCap 2D-DV
LSTM XGBoost LinReg LSTM XGBoost LinReg
\(F_x\) 10.3 (6.2) 17.3 (3.6) 17.6 (3.8) 12.7 (7.6) 18.6 (3.8) 18.2 (3.8)
\(F_y\) 7.1 (4.6) 13.4 (3.9) 19.0 (2.7) 10.4 (6.0) 20.2 (4.0) 23.5 (8.5)
\(F_z\) 4.3 (2.5) 11.1 (4.5) 11.0 (4.7) 10.7 (9.0) 19.8 (6.4) 18.6 (6.4)