Perceptual blur and ringing metrics: application to JPEG2000

By Pina Marziliano, Frédéric Dufaux, Stefan Winkler, Touradj Ebrahimi. Signal processing. Image communication, 2004

573 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/a39eb85c728c8fad3365ffd78b92bdb48fee82b3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Xiaokang Yang
3. A. Bovik
4. Ming-Hsuan Yang
5. Luc Van Gool
6. Kede Ma
7. Zhou Wang
8. Guangtao Zhai

Citations in the corpus

• Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wild (2020, methodology): “Early attempts at BIQA were tailored to specific synthetic distortions, such as JPEG compression [18] and JPEG2000 compression [19].”
• No-reference quality assessment using natural scene statistics: JPEG2000 (2005, methodology): “As a comparison, we can compare the performance of our algorithm against a no-reference perceptual blur metric presented in [19].”
• Reduced-reference image quality assessment using a wavelet-domain natural image statistic model (2005, background): “(7), but also provides a useful feature that when there is no distortion between the original and received images (which implies p(x) = q(x) for all x), both the targeted distortion measure d(p‖q) and estimated distortion measure d̂(p‖q) are exactly zero.”
• Perceptual visual quality metrics: A survey (2011, methodology): “A metric was devised in [101] for combining both blur and ringing measures in JPEG200 coded images.”
• dipIQ: Blind Image Quality Assessment by Learning-to-Rank Discriminable Image Pairs (2017, background): “The edge spread can be used to detect blurring [38], [39], and the intensity variance in smooth regions close to edges can indicate ringing artifacts [28].”
• Unsupervised feature learning framework for no-reference image quality assessment (2012, background): “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].”
• Reduced-Reference Image Quality Assessment Using Divisive Normalization-Based Image Representation (2009, background): “…the IQA task becomes very difficult, and almost all existing no-reference IQA metrics were designed for one or a set of predefined specific distortion types (such as blocking [2]–[5] and blurring [5] in JPEG; and ringing [6], blurring [6] and wavelet quantization effect [7], [8] in JPEG2000).”
• A No-Reference Objective Image Sharpness Metric Based on the Notion of Just Noticeable Blur (JNB) (2009, methodology): “Note that the sharpness metric of [7] is also referred to as the perceptual blur metric in the literature but it does not incorporate HVS-based perceptual weighting.”
• Waterloo Exploration Database: New Challenges for Image Quality Assessment Models (2017, unknown): “[29], [41].”
• Seven Challenges in Image Quality Assessment: Past, Present, and Future Research (2013, methodology): “A similar method can also be seen in [254].”
• Blind Image Quality Assessment Using a General Regression Neural Network (2011, background): “In [ 23 ], a more specific (blur only) NR algorithm was developed also for JPEG 2000 images.”
• A no-reference metric for evaluating the quality of motion deblurring (2013, methodology): “Methods have been proposed for measuring ringing artifacts [Marziliano et al. 2004], but most of them are only applicable to small scale ringing caused by lossy compression such as JPEG compression, and/or are full-reference metrics.”