Deep Learning for Blind Image Quality Assessment

In this work we investigate the use of deep learning for distortion-generic blind image quality assessment. We report on different design choices, ranging from the use of features extracted from pre-trained Convolutional Neural Networks (CNNs) as a generic image description, to the use of features extracted from a CNN fine-tuned for the image quality task. Our best proposal, named DeepBIQ, estimates the image quality by average-pooling the scores predicted on multiple sub-regions of the original image. The score of each sub-region is computed using a Support Vector Regression (SVR) machine taking as input features extracted using a CNN fine-tuned for category-based image quality assessment. Experimental results on the LIVE In the Wild Image Quality Challenge Database show that DeepBIQ outperforms the state-of-the-art methods compared, including those based on deep learning, having a Linear Correlation Coefficient (LCC) with human subjective scores of almost 0.91. These results are further confirmed also on four benchmark databases of synthetically distorted images: LIVE, CSIQ, TID2008 and TID2013. Furthermore, in most of the cases, the quality score predictions of DeepBIQ are closer to the average observer than those of a generic human observer.

Click here to download an extended version of the paper with additional material



On the use of deep learning for blind image quality assessment
(Simone Bianco, Luigi Celona, Paolo Napoletano, Raimondo Schettini) In Signal, Image and Video Processing, volume 12, number 2, pp. 355-362, 2018.

 author = {Bianco, Simone and Celona, Luigi and Napoletano, Paolo and Schettini, Raimondo},
 year = {2018},
 month = {2},
 year = {2018},
 pages = {355-362},
 title = {On the use of deep learning for blind image quality assessment},
 volume = {12},
 number = {2},
 journal = {Signal, Image and Video Processing},
 url = {},
 pdf = {},
 doi = {10.1007/s11760-017-1166-8},
 issn = {1863-1711},
 projectref = {}}