EURASIP Journal on Image and Video Processing

Table 3 Confusion matrix for the test set of Dataset-A

From: Time-dependent bag of words on manifolds for geodesic-based classification of video activities towards assisted living and healthcare

		Predicted class				Accuracy (%)
		Eating	Drinking	Using	Reading	Falling	Lying	Walking	Sitting
				laptop		down	down		down
True class	Eating	49	2	0	0	0	0	0	3	90.74
	Drinking	3	46	1	3	0	0	0	1	85.19
	Using laptop	1	1	45	3	0	0	0	2	86.54
	Reading	0	2	4	46	0	0	0	0	88.46
	Falling down	0	0	0	0	50	4	0	0	92.59
	Lying down	0	0	0	0	4	49	0	0	92.45
	Walking	1	0	3	4	0	0	67	0	89.33
	Sitting down	2	0	2	2	0	0	0	69	92.00

Back to article page