UNIQUE: Unsupervised Image Quality Estimation
This addresses image quality assessment for applications like image processing, but it is incremental as it builds on existing unsupervised and sparse representation methods.
The paper tackles the problem of estimating perceived image quality without supervision by using sparse representations from generic image databases, and results show that UNIQUE is a top-performing estimator in accuracy, consistency, linearity, and monotonic behavior on multiple benchmark databases.
In this paper, we estimate perceived image quality using sparse representations obtained from generic image databases through an unsupervised learning approach. A color space transformation, a mean subtraction, and a whitening operation are used to enhance descriptiveness of images by reducing spatial redundancy; a linear decoder is used to obtain sparse representations; and a thresholding stage is used to formulate suppression mechanisms in a visual system. A linear decoder is trained with 7 GB worth of data, which corresponds to 100,000 8x8 image patches randomly obtained from nearly 1,000 images in the ImageNet 2013 database. A patch-wise training approach is preferred to maintain local information. The proposed quality estimator UNIQUE is tested on the LIVE, the Multiply Distorted LIVE, and the TID 2013 databases and compared with thirteen quality estimators. Experimental results show that UNIQUE is generally a top performing quality estimator in terms of accuracy, consistency, linearity, and monotonic behavior.