Table 2.

Comparison of different models on different indicators.

Method	Dataset
	Swimming Technique Datasets				ImageNet Datasets				Kinetics Datasets				Sports-1M Dataset
	Parameters(M)	Flops(G)	Inference Time(ms)	Trainning Time(s)	Parameters(M)	Flops(G)	Inference Time(ms)	Trainning Time(s)	Parameters(M)	Flops(G)	Inference Time(ms)	Trainning Time(s)	Parameters(M)	Flops(G)	Inference Time(ms)	Trainning Time(s)
Wang et al.	388.10	375.78	316.33	333.84	392.92	387.56	322.59	231.39	319.72	254.78	205.01	370.22	316.82	281.96	240.08	290.39
Aust et al.	389.89	361.01	249.66	308.66	366.67	210.75	399.53	381.73	312.27	339.02	223.11	210.46	317.75	361.67	223.75	358.71
Biewe et al.	300.70	379.74	372.24	258.53	351.51	261.25	242.28	264.86	246.54	379.88	376.79	225.63	329.81	396.55	358.34	265.83
Sant et al.	349.32	283.54	274.55	305.18	280.76	229.60	214.61	349.32	298.49	398.13	358.91	200.38	203.88	395.75	311.97	302.91
Cabr et al.	215.34	335.34	361.48	221.11	339.58	237.88	200.41	301.59	264.36	233.56	369.14	235.64	389.59	314.20	227.29	309.49
Ma et al.	326.29	305.51	303.25	371.38	304.10	232.16	334.90	380.34	227.60	335.66	215.48	390.91	395.46	346.65	209.33	355.87
Ours	177.00	222.90	158.58	114.97	192.73	133.35	108.83	197.33	191.45	122.30	208.18	128.27	143.93	118.94	202.95	126.93