EURASIP Journal on Image and Video Processing

Table 4 Recognition accuracy (%) of various feature combinations on HMDB-LQ and HMDB-MQ subsets

From: Exploiting textures for better action recognition in low-quality videos

Method	HMDB-LQ	HMDB-MQ	Method	HMDB-LQ	HMDB-MQ
STIP (baseline)	21.71	23.68	iDT (Baseline)	23.88	41.43
STIP+LBP-TOP	20.80	24.28	iDT+LBP-TOP	30.34	43.11
STIP+LPQ-TOP	23.89	28.36	iDT+LPQ-TOP	30.96	42.97
STIP+BSIF-TOP	32.46	37.14	iDT+BSIF-TOP	37.80	45.96
STIP+deep object [35]	34.57	42.48	iDT+deep object [35]	36.51	44.80

Back to article page