Visual Interaction Force Estimation Based on Time-Sensitive Dual-Resolution Learning Network

Table 4

The influence of different hyperparameters on the performance of our method.

Model	RMSE	MAE	MSE
, ,	0.0472	0.0288	0.0022	0.9733
, , (baseline)	0.0523	0.0332	0.0027	0.9650
, , ,	0.0704	0.0446	0.0049	0.9446
, , ,	0.0596	0.0404	0.0035	0.9567
, , ,	0.0455	0.0284	0.0021	0.9739
, , ,	0.0568	0.0346	0.0032	0.9639
, , ,	0.0401	0.0272	0.0016	0.9794
, , ,	0.0359	0.0228	0.0013	0.9841
, , ,	0.0503	0.0366	0.0025	0.9701