scispace - formally typeset
Open AccessJournal ArticleDOI

A Convolutional Neural Network Approach for Objective Video Quality Assessment

P. Le Callet, +2 more
- 01 Sep 2006 - 
- Vol. 17, Iss: 5, pp 1316-1327
Reads0
Chats0
TLDR
The proposed approach solves some problems inherent to objective metrics that should predict subjective quality score obtained using the single stimulus continuous quality evaluation (SSCQE) method and relies on the use of a convolutional neural network that allows a continuous time scoring of the video.
Abstract: 
This paper describes an application of neural networks in the field of objective measurement method designed to automatically assess the perceived quality of digital videos. This challenging issue aims to emulate human judgment and to replace very complex and time consuming subjective quality assessment. Several metrics have been proposed in literature to tackle this issue. They are based on a general framework that combines different stages, each of them addressing complex problems. The ambition of this paper is not to present a global perfect quality metric but rather to focus on an original way to use neural networks in such a framework in the context of reduced reference (RR) quality metric. Especially, we point out the interest of such a tool for combining features and pooling them in order to compute quality scores. The proposed approach solves some problems inherent to objective metrics that should predict subjective quality score obtained using the single stimulus continuous quality evaluation (SSCQE) method. This latter has been adopted by video quality expert group (VQEG) in its recently finalized reduced referenced and no reference (RRNR-TV) test plan. The originality of such approach compared to previous attempts to use neural networks for quality assessment, relies on the use of a convolutional neural network (CNN) that allows a continuous time scoring of the video. Objective features are extracted on a frame-by-frame basis on both the reference and the distorted sequences; they are derived from a perceptual-based representation and integrated along the temporal axis using a time-delay neural network (TDNN). Experiments conducted on different MPEG-2 videos, with bit rates ranging 2-6 Mb/s, show the effectiveness of the proposed approach to get a plausible model of temporal pooling from the human vision system (HVS) point of view. More specifically, a linear correlation criteria, between objective and subjective scoring, up to 0.92 has been obtained on a set of typical TV videos

read more

Content maybe subject to copyright    Report

HAL Id: hal-00287426
https://hal.archives-ouvertes.fr/hal-00287426
Submitted on 11 Jun 2008
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
A Convolutional Neural Network Approach for
Objective Video Quality Assessment
Patrick Le Callet, Christian Viard-Gaudin, Dominique Barba
To cite this version:
Patrick Le Callet, Christian Viard-Gaudin, Dominique Barba. A Convolutional Neural Network Ap-
proach for Objective Video Quality Assessment. IEEE Transactions on Neural Networks, Institute
of Electrical and Electronics Engineers, 2006, 17 (5), pp.1316- 1327. �10.1109/TNN.2006.879766�.
�hal-00287426�

 !"# $%&'(
 )  *+ )   #, - .
./
./0
11"
 !,2") 
!&3%3'
4453% 1 6,$75!8
6055-43%95343
87055-43%95-5-
,0:);
1) *+1..;
<=)>1
A Convolutional Neural Network Approach for Objective Video Quality Assessment
?

6....;*+;;=)
 /     .)  ; / ) 6
//      =/   . ) .7  
/=)$))..
+6/;*+;;/
;/.7.6;.../
.;;/**+
;*+   7 ;  ;   ,. * .  
; ;/;./  .
 6 .. .. )  .   =)  
.=)//
)@$$>,A  6.>,@>,7.
.A;B 6@; ;A.
6 / ;  .. . .) .    *+ ;
;)*+@ A*
  / ;  ) <=) ;  7   ;;
;);..
 .  / /  . 7 /  6  
*+ @6 A ,7.   ;; #,- ) *  
// ; -  % #C *  ;;) ;  .. ..  / 
.  ; . ./ ;   )  @D$A . ; )*
#.;*=)=)/.
3'-;.6)
Complete final manuscript

IEEE Transactions on Neural Networks
-
I Introduction
*)./..;.;
=;)*6.);
) ; . .$.+/ ;    
)=..;.;/
 )     /  # )   . ;
;;//.;
.;/)+;*)1;;
;))/);/./
. ; 6 ) /    . ; / ;;
;/;6.#,.
/#,.@#6$A*.7*
 #, .  ;       .B 
*1);./
.7 ;  #6$   ./  .E /
   .  .) ;; ) .  )    
      / #,- *  ; -  &
#C;;)@$6A?&?'#C;/;
)@D6AD*).+.)
 ,;;) * ;; )  .      * 
.   .)  ; /       
./*+/
*+...*/
;.#,-/
Citations
More filters
Journal ArticleDOI

Big Data Deep Learning: Challenges and Perspectives

Xue-wen Chen, +1 more
- 16 May 2014 - 
TL;DR: A brief overview of deep learning is provided, and current research efforts and the challenges to big data, as well as the future trends are highlighted.
Journal ArticleDOI

A survey of machine learning for big data processing

Junfei Qiu, +4 more
- 28 May 2016 - 
TL;DR: A literature survey of the latest advances in researches on machine learning for big data processing finds some promising learning methods in recent studies, such as representation learning, deep learning, distributed and parallel learning, transfer learning, active learning, and kernel-based learning.
Journal ArticleDOI

Intelligent fault diagnosis of rolling bearing using hierarchical convolutional network based health state classification

Chen Lu, +2 more
- 01 Apr 2017 - 
TL;DR: A novel deep architecture based bearing diagnosis method is proposed using cognitive computing theory, which introduces the advantages of image recognition and visual perception to bearing fault diagnosis by simulating the cognition process of the cerebral cortex.
Journal ArticleDOI

Image Quality Assessment Based on Multiscale Geometric Analysis

Xinbo Gao, +3 more
- 01 Jul 2009 - 
TL;DR: A novel framework for IQA to mimic the human visual system (HVS) by incorporating the merits from multiscale geometric analysis (MGA), contrast sensitivity function (CSF), and the Weber's law of just noticeable difference (JND) is developed.
Journal ArticleDOI

Reduced-Reference Image Quality Assessment Using Reorganized DCT-Based Image Representation

Lin Ma, +3 more
- 01 Aug 2011 - 
TL;DR: The proposed reduced-reference image quality assessment (IQA) outperforms the RR WNISM and even the full-reference (FR) quality metric PSNR and demonstrates that only a small number of RR features is sufficient for representing the image perceptual quality.
References
More filters
Journal ArticleDOI

Gradient-based learning applied to document recognition

Yann LeCun, +6 more
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Book

Neural networks for pattern recognition

Christopher M. Bishop
TL;DR: This is the first comprehensive treatment of feed-forward neural networks from the perspective of statistical pattern recognition, and is designed as a text, with over 100 exercises, to benefit anyone involved in the fields of neural computation and pattern recognition.
Book ChapterDOI

Neural Networks for Pattern Recognition

Suresh Kothari, +1 more
- 01 Jan 1993 - 
TL;DR: The chapter discusses two important directions of research to improve learning algorithms: the dynamic node generation, which is used by the cascade correlation algorithm; and designing learning algorithms where the choice of parameters is not an issue.
Journal ArticleDOI

Receptive fields, binocular interaction and functional architecture in the cat's visual cortex

David H. Hubel, +1 more
- 01 Jan 1962 - 
TL;DR: This method is used to examine receptive fields of a more complex type and to make additional observations on binocular interaction and this approach is necessary in order to understand the behaviour of individual cells, but it fails to deal with the problem of the relationship of one cell to its neighbours.
Book

Phoneme recognition using time-delay neural networks

Alex Waibel, +4 more
TL;DR: The authors present a time-delay neural network (TDNN) approach to phoneme recognition which is characterized by two important properties: using a three-layer arrangement of simple computing units, a hierarchy can be constructed that allows for the formation of arbitrary nonlinear decision surfaces, which the TDNN learns automatically using error backpropagation.
Related Papers (5)

Image quality assessment: from error visibility to structural similarity

Zhou Wang, +3 more
- 01 Apr 2004 - 

A new standardized method for objectively measuring video quality

M.H. Pinson, +1 more
- 27 Sep 2004 - 

Study of Subjective and Objective Quality Assessment of Video

Kalpana Seshadrinathan, +3 more
- 01 Jun 2010 - 

Multiscale structural similarity for image quality assessment

Zhou Wang, +2 more

A Statistical Evaluation of Recent Full Reference Image Quality Assessment Algorithms

Hamid R. Sheikh, +2 more
- 01 Nov 2006 - 
Frequently Asked Questions (1)
Q1. What contributions have the authors mentioned in the paper "A convolutional neural network approach for objective video quality assessment" ?

HAL this paper is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not.