Regression or classification? New methods to evaluate no-reference picture and video quality models

Abstract

Video and image quality assessment has long been projected as a regression problem, which requires predicting a continuous quality score given an input stimulus. However, recent efforts have shown that accurate quality score regression on real-world user-generated content (UGC) is a very challenging task. To make the problem more tractable, we propose two new methods -binary, and ordinal classification -as alternatives to evaluate and compare no-reference quality models at coarser levels. Moreover, the proposed new tasks convey more practical meaning on perceptually optimized UGC transcoding, or for preprocessing on media processing platforms. We conduct a comprehensive benchmark experiment of popular no-reference quality models on recent in-the-wild picture and video quality datasets, providing reliable baselines for both evaluation methods to support further studies. We hope this work promotes coarse-grained perceptual modeling and its applications to efficient UGC processing.

Quadratic fitting model in no-reference image quality assessment

“…NSS features found in NR-IQA metrics should be able to identify traits susceptible to particular sorts of distortions. There are other approaches and features in addition to GGD [20], where other distributions can be useful as well. Among the methods there is also dmos distribution modeling, where greater length of feature vector is used or where novel classifiers, distortion and image type combinations are tested, etc.…”

Deep neural networks for evaluation of specular microscopy images of the corneal endothelium with Fuchs' dystrophy

“…However, these methods rely heavily on extensive manually annotated datasets. [5][6][7][8][9] Because of the above, unsupervised learning methods have emerged to decrease the cost and time of manual image labeling. [10][11][12][13][14] The latter, use strategies such as; clustering, transfer learning, self-supervision and other coding tools.…”