Table 1. Experimental results of synthetic distortion dataset.

Datasets	LIVE [25]		CSIQ [26]		TID2013 [27]		Weight average
Methods	SROCC	PLCC	SROCC	PLCC	SROCC	PLCC	SROCC	PLCC
PSNR	0.876	0.872	0.806	0.800	0.636	0.706	0.773	0.793
SSIM [19]	0.913	0.945	0.834	0.861	0.775	0.691	0.841	0.833
FSIMc [23]	0.963	0.960	0.913	0.919	0.802	0.877	0.893	0.919
BRISQUE [30]	0.939	0.942	0.775	0.817	0.572	0.651	0.762	0.803
CORNIA [31]	0.942	0.943	0.714	0.781	0.549	0.613	0.735	0.779
IL-NIQE [32]	0.902	0.908	0.821	0.865	0.521	0.648	0.748	0.807
CNN [33]	0.956	0.954	0.683	0.754	0.558	0.653	0.732	0.787
HOSA [33]	0.948	0.949	0.781	0.841	0.688	0.764	0.806	0.851
FRIQUEE [35]	0.940	0.944	0.835	0.874	0.680	0.753	0.818	0.857
RANK [14]	0.981	0.982	0.861	0.893	0.780	0.793	0.874	0.889
DMIR-IQA [36]	0.967	0.971	0.823	0.881	0.796	0.821	0.862	0.912
MMMNet [37]	0.970	0.970	0.924	0.937	0.832	0.853	0.908	0.920
AIGQA [38]	0.960	0.957	0.927	0.952	0.871	0.893	0.919	0.934
DB-CNN [15]	0.968	0.971	0.946	0.959	0.816	0.865	0.910	0.931
Deep-FL [39]	0.972	0.978	0.930	0.946	0.858	0.876	0.891	0.907
CaHDC [40]	0.965	0.964	0.903	0.914	0.816	0.865	0.895	0.914
NSSADNN [41]	0.986	0.984	0.893	0.927	0.844	0.910	0.907	0.940
BaseLine	0.950	0.954	0.876	0.905	0.712	0.756	0.846	0.872
Ours	0.976	0.980	0.942	0.954	0.895	0.903	0.940	0.945