Two-layer similarity fusion model for cover song identification

Chen, Ning; Li, Mingyu; Xiao, Haidong

doi:10.1186/s13636-017-0108-2

EURASIP Journal on Audio, Speech, and Music Processing

Table 6 Classification efficiency achieved by different descriptor combinations

From: Two-layer similarity fusion model for cover song identification

	Combination of HPCP and MLD				Combination of BSC and CPCP
		TNR	CA (%)	ACA		TNR	CA (%)	ACA
DB801	2L-Best1	0.468	99.7718	0.7340	2L-Best1	0.417	99.7715	0.7081
	2L-HPCP-Best1	0.420	99.7692	0.7100	2L-BSC-Best1	0.401	99.7662	0.7003
	1L-HPCP-QD [32]	0.420	99.7672	0.7100	1L-BSC-QD	0.401	99.7662	0.7003
	1L-MLD-QD	0.157	99.6891	0.5785	1L-CPCP-QD	0.063	99.6604	0.5316
	[26]	0.359	99.7541	0.6795	[26]	0.359	99.7541	0.6795
	[24]	0.456	98.8105	0.7230	[24]	0.456	98.8105	0.7230
	PSO	0.279	99.7289	0.6395	PSO	0.242	99.7218	0.6208
	[31]	0.114	99.6749	0.5570	[31]	0.089	99.6689	0.5446
DB799	2L-Best1	0.486	99.7729	0.7425	2L-Best1	0.370	99.7778	0.6847
	2L-HPCP-Best1	0.431	99.7710	0.7155	2L-BSC-Best1	0.355	99.7450	0.6774
	1L-HPCP-QD [32]	0.431	99.7710	0.7155	1L-BSC-QD	0.355	99.7450	0.6774
	1L-MLD-QD	0.166	99.7020	0.5830	1L-CPCP-QD	0.058	99.6741	0.5291
	[26]	0.347	99.7541	0.6735	[26]	0.347	99.7541	0.6735
	[24]	0.408	98.7142	0.6985	[24]	0.408	98.7142	0.6985
	PSO	0.259	99.7360	0.6295	PSO	0.288	99.6281	0.6435
	[31]	0.112	99.6931	0.5560	[31]	0.066	99.6772	0.5332
DB802	2L-Best1	0.464	99.7709	0.7320	2L-Best1	0.372	99.7565	0.6859
	2L-HPCP-Best1	0.430	99.7755	0.7150	2L-BSC-Best1	0.366	99.7562	0.6832
	1L-HPCP-QD [32]	0.430	99.7755	0.7150	1L-BSC-QD	0.366	99.7562	0.6832
	1L-MLD-QD	0.163	99.7014	0.5815	1L-CPCP-QD	0.060	99.6762	0.5299
	[26]	0.340	99.7572	0.6700	[26]	0.340	99.7572	0.6700
	[24]	0.407	99.1071	0.7000	[24]	0.407	99.1071	0.7000
	PSO	0.429	98.5529	0.7080	PSO	0.202	99.7289	0.6395
	[31]	0.101	99.6908	0.5505	[31]	0.060	99.6771	0.5299
DB962	2L-Best1	0.498	99.8453	0.7490	2L-Best1	0.383	99.8302	0.6915
	2L-HPCP-Best1	0.443	99.8440	0.7215	2L-BSC-Best1	0.373	99.8295	0.6865
	1L-HPCP-QD [32]	0.443	99.8440	0.7215	1L-BSC-QD	0.373	99.8295	0.6865
	1L-MLD-QD	0.186	99.7990	0.5930	1L-CPCP-QD	0.079	99.7796	0.5397
	[26]	0.377	99.8354	0.6885	[26]	0.377	99.8354	0.6885
	[24]	0.423	99.1193	0.7080	[24]	0.423	99.1193	0.7080
	PSO	0.740	71.7942	0.7290	PSO	0.239	99.8040	0.6194
	[31]	0.140	99.7926	0.5700	[31]	0.085	99.7814	0.5424

Notes: For TNR, CA, and ACA, the larger the value is, the better performance is obtained

Back to article page