Driver’s Face Pose Estimation Using Fine-Grained Wi-Fi Signals for Next-Generation Internet of Vehicles

Table 6

Classification accuracy (%) of KSVM with varying values.

K value	Gesture class
	RS	LS	RM	LM	RA	LI	CM	HD	HR	HL
	87.5	88.2	88.4	88.7	89.1	86.9	89.2	88.1	90.3	89.7
	89.2	88.9	90.2	89.7	88.5	89.3	90.6	87.7	90.4	90.4
	90.4	91.1	92.6	92.5	89.3	90.7	91.4	89.9	92.2	92.1
	90.5	89.3	91.7	92.1	89.7	90.3	89.9	90.2	91.6	91.0
	89.3	91.0	90.5	90.6	89.4	89.6	90.1	88.9	90.6	90.1
	89.6	91.3	91.7	88.5	88.9	90.3	90.7	88.4	91.0	90.4
	88.3	90.7	90.6	91.1	89.6	90.1	88.9	89.0	90.5	88.6
	90.5	89.1	89.5	90.2	89.8	89.6	88.7	89.0	90.3	90.6
	87.5	90.0	88.7	87.3	88.1	87.7	89.3	88.4	89.0	87.6
	87.3	89.4	89.1	88.2	89.0	86.9	88.1	87.5	86.8	88.2
	89.1	87.9	90.0	88.8	89.5	86.7	88.3	87.5	88.1	86.2
	86.3	89.5	89.1	89.0	87.7	86.5	88.2	86.9	90.0	89.5
	86.5	86.3	87.6	89.0	86.3	88.9	88.2	90.1	86.2	87.8