Learning to rank for blind image quality

Blind image quality assessment (BIQA) aims to predict perceptual image quality scores without access to reference images. State-of-the-art BIQA methods typically require subjects to score a large number of images to train a robust model. However, the acquisition of image quality scores has several limitations: 1) scores are not precise, because subjects are usually uncertain about which score most precisely represents the perceptual quality of a given image; 2) subjective judgments of quality may be biased by image content; 3) the quality scales between different distortion categories are inconsistent, because images corrupted by different types of distortion are evaluated independently in subjective experiments; and 4) it is challenging to obtain a large scale database, or to extend existing databases, because of the inconvenience of collecting sufficient images associated with different kinds of distortion that have diverse levels of degradation, training the subjects, conducting subjective experiments, and realigning human quality evaluations. To combat these limitations, this paper explores and exploits preference image pairs (PIPs) such as “the quality of image Ia is better than that of image Ib” for training a robust BIQA model. The preference label, representing the relative quality of two images, is generally precise and consistent, and is not sensitive to image content, distortion type, or subject identity; such PIPs can be generated at very low cost. The proposed BIQA method is one of learning to rank. We first formulate the problem of learning the mapping from the image features to the preference label as one of classification. In particular, we investigate the utilization of a multiple kernel learning algorithm based on group lasso (MKLGL) to provide a solution. A simple but effective strategy to estimate perceptual image quality scores is then presented. Experiments show that the proposed BIQA method is highly effective and achieves comparable performance