Table 4.

Confusion matrix on the test set of the VGG19 network with weights initialized on ImageNet, and trained using the mean square error loss function and the Adam optimizer

	W	N	SN	SH	S	P	IF	BH	F	FA
W	96.1	0.3	0.0	0.2	2.0	0.0	0.0	0.0	0.0	1.3
N	0.1	95.5	2.6	0.5	0.0	0.3	0.0	0.8	0.1	0.0
SN	0.4	0.5	88.3	6.7	0.0	2.1	0.1	0.8	1.1	0.0
SH	1.0	0.9	5.4	77.4	0.2	2.8	0.6	8.8	2.0	1.0
S	8.2	0.0	0.0	0.4	88.2	0.0	0.0	0.0	0.0	3.2
P	0.1	0.1	0.1	0.3	0.0	90.9	2.4	4.0	0.1	1.9
IF	0.7	0.6	0.0	0.6	0.0	9.6	83.3	4.3	0.8	0.0
BH	0.5	0.3	0.5	4.8	0.9	1.9	0.6	86.6	1.8	2.0
F	0.0	0.0	0.0	0.9	0.1	1.1	1.8	0.3	95.8	0.0
FA	1.1	0.4	0.1	0.0	0.4	0.7	0.0	2.0	0.0	95.4

Results are referred to the single frame classifier, i.e., sliding windows size equal to 1