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

Table 5

Comparison results of ablation experiments.

Method	Interactive scene pathway	Interactive action pathway	Lightweight network design	Hybrid data augmentation		Inference time (s)
(1)	√				0.9268
(2)		√			0.8224
(3)	√	√			0.9613
(4)	√	√		√	0.9725
(5)	√	√	√	√	0.9833