Skip to main content

Table 1 Explicit NR-IQA algorithm performance comparison

From: Explicit-implicit dual stream network for image quality assessment

Algorithm Parameters No. of Indies on LIVE   Training Year
  used parameters SROCC PLCC method  
BIQI GGD(α,θ2) 18 0.8271 0.8417 SVM 2010
DIIVINE GGD(α,θ2) 88 0.9160 0.9170 SVM 2011
BLIINDS-II DCT 16 0.9124 0.9164 Model 2012
BRISQUE \(GGD\left (\alpha,\theta ^{2}\right),AGGD\left (\eta,\nu,\theta _{l}^{2},\theta _{r}^{2}\right)\) 36 0.9395 0.9424 SVM 2012
DESIQUE \(GGD\left (\alpha _{i},\theta _{i}^{2}\right)\) 60 0.9437 0.9465 SVM 2013
NIQE \(GGD\left (\alpha,\theta ^{2}\right),AGGD\left (\eta,\nu,\theta _{l}^{2},\theta _{r}^{2}\right)\) 36 0.9153 0.9147 Model 2013
SSEQ entropy 12 0.9348 0.9383 SVM 2014
IL-NIQE \(GGD\left (\alpha,\theta ^{2}\right),AGGD\left (\eta,\nu,\theta _{l}^{2},\theta _{r}^{2}\right)\) 468 0.9025 0.9085 Model 2015