28th Picture Coding Symposium 2010
DOI: 10.1109/pcs.2010.5702578
|View full text |Cite
|
Sign up to set email alerts
|

A reduced-reference metric based on the interest points in color images

Abstract: Il est utile de pouvoir quantifier les dégradations perçues afin de juger l'intérêt de nouveaux algorithmes de compression, tatouage ou des techniques de transmission. Nous proposons d'évaluer les performances d'un lot de métriques de qualité d'images couleurs en visant la corrélation avec le jugement humain. Le but escompté est de faciliter le choix d'une métrique parmi les nombreuses disponibles, en fournissant des scores de performance standards et exhaustifs. Pour cette étude quatre bases de données d'imag… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
5
0

Year Published

2011
2011
2020
2020

Publication Types

Select...
5
2
1

Relationship

1
7

Authors

Journals

citations
Cited by 9 publications
(5 citation statements)
references
References 21 publications
0
5
0
Order By: Relevance
“…The biggest difficulty in designing RR metrics is the extraction of suitable and descriptive features from the reference video sequence in order to have sufficient data for an accurate video quality prediction [18]. Tao et al [19] proposed a relative video quality metric, rPSNR, that can be computed without parsing or decoding the transmitted video, and without any knowledge of video characteristics.…”
Section: Reduced-reference Metricsmentioning
confidence: 99%
“…The biggest difficulty in designing RR metrics is the extraction of suitable and descriptive features from the reference video sequence in order to have sufficient data for an accurate video quality prediction [18]. Tao et al [19] proposed a relative video quality metric, rPSNR, that can be computed without parsing or decoding the transmitted video, and without any knowledge of video characteristics.…”
Section: Reduced-reference Metricsmentioning
confidence: 99%
“…This strategy allows to exploit a small part of the reference for quality measurement purposes. QIP [9] belongs to the last category. It is based on the ability of the interest points to predict a variation in the image, depending on the object's saliency.…”
Section: Decoding Strategy With Qipmentioning
confidence: 99%
“…The aim is to determine whether the interest points can be used to predict salient informations on an image like the HVS does. This can help for several applications like quality assessment, 10 simplified saliency maps construction, 11 . .…”
Section: Introductionmentioning
confidence: 99%