Disentangling Image Distortions in Deep Feature Space

02/26/2020
by   Simone Bianco, et al.
6

Previous literature suggests that perceptual similarity is an emergent property shared across deep visual representations. Experiments conducted on a dataset of human-judged image distortions have proven that deep features outperform, by a large margin, classic perceptual metrics. In this work we take a further step in the direction of a broader understanding of such property by analyzing the capability of deep visual representations to intrinsically characterize different types of image distortions. To this end, we firstly generate a number of synthetically distorted images by applying three mainstream distortion types to the LIVE database and then we analyze the features extracted by different layers of different Deep Network architectures. We observe that a dimension-reduced representation of the features extracted from a given layer permits to efficiently separate types of distortions in the feature space. Moreover, each network layer exhibits a different ability to separate between different types of distortions, and this ability varies according to the network architecture. As a further analysis, we evaluate the exploitation of features taken from the layer that better separates image distortions for: i) reduced-reference image quality assessment, and ii) distortion types and severity levels characterization on both single and multiple distortion databases. Results achieved on both tasks suggest that deep visual representations can be unsupervisedly employed to efficiently characterize various image distortions.

READ FULL TEXT

page 1

page 3

research
05/16/2023

Deep Ensembling for Perceptual Image Quality Assessment

Blind image quality assessment is a challenging task particularly due to...
research
01/11/2018

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

While it is nearly effortless for humans to quickly assess the perceptua...
research
12/21/2020

A Shift-insensitive Full Reference Image Quality Assessment Model Based on Quadratic Sum of Gradient Magnitude and LOG signals

Image quality assessment that aims at estimating the subject quality of ...
research
05/22/2017

Learning multiple visual domains with residual adapters

There is a growing interest in learning data representations that work w...
research
12/14/2017

RAN4IQA: Restorative Adversarial Nets for No-Reference Image Quality Assessment

Inspired by the free-energy brain theory, which implies that human visua...
research
12/03/2018

Attention-based Adaptive Selection of Operations for Image Restoration in the Presence of Unknown Combined Distortions

Many studies have been conducted so far on image restoration, the proble...
research
11/30/2018

Hybrid Distortion Aggregated Visual Comfort Assessment for Stereoscopic Image Retargeting

Visual comfort is a quite important factor in 3D media service. Few rese...

Please sign up or login with your details

Forgot password? Click here to reset