A Grassmann graph embedding framework for gait analysis

EURASIP Journal on Advances in Signal Processing

Table 6 Evaluating the effect of speed variations using rank-1 recognition rate (%)

		Benchmark		Global method				Local method				Adaptive method
		Score fusion	MSR[44]	SM-CC	SM-Proj	HK-CC	HK-Proj	SM-CC	SM-Proj	HK-CC	HK-Proj	SR-CC	SR-Proj
Unsupervised GGE	TR, all speeds	97.06	100	100	94.12	100	94.12	100	100	100	100	100	100
	TT, all speeds	97.06	100	100	94.12	100	94.12	100	100	100	100	100	100
	TR 1, 2, 3	10.29	26.47	35.29	2.94	35.29	2.94	35.29	2.94	35.29	2.94	44.71	26.25
	TT 4, 5, 6	10.29	26.47	35.29	2.94	35.29	2.94	35.29	2.94	35.29	2.94	44.71	26.25
	TR 1, 2	5.88	14.71	11.76	11.76	11.76	11.76	11.76	11.76	11.76	11.76	23.53	23.53
	TT 5, 6	5.88	14.71	11.76	11.76	11.76	11.76	11.76	11.76	11.76	11.76	23.53	23.53
Supervised GGE	TR, all speeds	97.06	100	100	94.12	100	94.12	100	94.12	100	94.12	100	100
	TT, all speeds	97.06	100	100	94.12	100	94.12	100	94.12	100	94.12	100	100
	TR 1, 2, 3	10.29	26.47	70.59	11.76	67.64	17.65	70.59	11.76	70.59	11.76	70.65	17.65
	TT 4, 5, 6	10.29	26.47	70.59	11.76	67.64	17.65	70.59	11.76	70.59	11.76	70.65	17.65
	TR 1, 2	5.88	14.71	29.41	11.76	29.41	8.82	29.41	11.76	35.29	14.71	35.71	8.82
	TT 5, 6	5.88	14.71	29.41	11.76	29.41	8.82	29.41	11.76	35.29	14.71	35.71	8.82