Repository logo
 

Dataset and metrics for predicting local visible differences

Accepted version
Peer-reviewed

Type

Article

Change log

Authors

Wolski, K 
Giunchi, D 
Ye, N 
Didyk, P 
Myszkowski, K 

Abstract

jats:pA large number of imaging and computer graphics applications require localized information on the visibility of image distortions. Existing image quality metrics are not suitable for this task as they provide a single quality value per image. Existing visibility metrics produce visual difference maps, and are specifically designed for detecting just noticeable distortions but their predictions are often inaccurate. In this work, we argue that the key reason for this problem is the lack of large image collections with a good coverage of possible distortions that occur in different applications. To address the problem, we collect an extensive dataset of reference and distorted image pairs together with user markings indicating whether distortions are visible or not. We propose a statistical model that is designed for the meaningful interpretation of such data, which is affected by visual search and imprecision of manual marking. We use our dataset for training existing metrics and we demonstrate that their performance significantly improves. We show that our dataset with the proposed statistical model can be used to train a new CNN-based metric, which outperforms the existing solutions. We demonstrate the utility of such a metric in visually lossless JPEG compression, super-resolution and watermarking.</jats:p>

Description

Keywords

Visual perception, visual difference predictor, visual metric, distortion visibility, image quality, data-driven metric, dataset, convolutional neural network

Journal Title

ACM Transactions on Graphics

Conference Name

Journal ISSN

0730-0301
1557-7368

Volume Title

37

Publisher

Association for Computing Machinery (ACM)
Sponsorship
European Research Council (725253)
European Commission Horizon 2020 (H2020) Marie Sklodowska-Curie actions (765911)