A No-Reference Modular Video Quality Prediction Model for H.265/HEVC and VP9 Codecs on a Mobile Device

Table 25

Weight and bias value for input layer.

Weight								Bias
PL	J	T	ARS	BR	FR	R	Codec
0.769	0.073	0.644	−0.35	0.540	−0.20	−0.53	0.228	−2.18
0.099	0.342	1.006	0.182	0.933	0.682	1.657	0.101	1.44
1.085	0.183	0.256	0.383	−0.80	−0.10	−0.01	−0.28	−1.57
1.117	0.293	0.194	0.232	1.037	0.478	0.786	0.828	−1.13
−0.90	−0.31	0.238	0.900	−0.44	−0.24	0.847	−0.36	1.13
0.775	0.508	0.893	0.537	0.880	1.801	0.644	0.025	−0.23
0.106	0.351	0.694	−0.97	1.182	−0.53	0.566	0.002	−0.28
0.505	1.085	0.276	0.462	1.710	0.161	1.542	0.094	0.34
0.169	1.144	0.094	1.648	−0.59	1.476	0.571	−0.04	0.21
0.329	0.742	0.133	0.085	0.515	0.324	0.473	1.206	0.42