Unsupervised feature learning framework for no-reference image quality assessment

In this paper, we present an efficient general-purpose objective no-reference (NR) image quality assessment (IQA) framework based on unsupervised feature learning. The goal is to build a computational model to automatically predict human perceived image quality without a reference image and without knowing the distortion present in the image. Previous approaches for this problem typically rely on hand-crafted features which are carefully designed based on prior knowledge. In contrast, we use raw-image-patches extracted from a set of unlabeled images to learn a dictionary in an unsupervised manner. We use soft-assignment coding with max pooling to obtain effective image representations for quality estimation. The proposed algorithm is very computationally appealing, using raw image patches as local descriptors and using soft-assignment for encoding. Furthermore, unlike previous methods, our unsupervised feature learning strategy enables our method to adapt to different domains. CORNIA (Codebook Representation for No-Reference Image Assessment) is tested on LIVE database and shown to perform statistically better than the full-reference quality measure, structural similarity index (SSIM) and is shown to be comparable to state-of-the-art general purpose NR-IQA algorithms.

1. Introduction

Our work addresses the problem of no-reference (NR) objective image quality assessment (IQA) on natural scene images. The goal is to build a computational model to predict human perceived image quality, accurately and automatically without access to reference images [1], [11], [13], [14], [17], [18], [22], [23], [26]. In the past several decades, there has been an increased interest in objective IQA due to the tremendous growth in the use of digital images for representing and communicating information. Objective image quality measures have been used in a wide range of computer vision and image processing applications. For example, image processing and transmission systems may have their parameters be adjusted according to the image quality [9]; image retrieval systems can use quality as an attribute to rank images and image processing algorithms may use image quality measures for evaluation.