TY  - JOUR
T1  - Full reference video quality evaluation using foveated vision and multiple fixation points
AU  - Vranješ, Mario
AU  - Rimac-drlje, Snježana
AU  - Vranješ, Denis
PY  - 2019
DA  - July
Y2  - 2019
DO  - 10.22399/ijcesen.477034
JF  - International Journal of Computational and Experimental Science and Engineering
JO  - IJCESEN
PB  - İskender AKKURT
WT  - DergiPark
SN  - 2149-9144
SP  - 94
EP  - 99
VL  - 5
IS  - 2
LA  - en
AB  - Invideo applications it is necessary to continuously measure the video qualityperceived by the end-user. Thus it is desirable to know which parts of videoframe, i.e. which contents, attract viewers’ attention. If this information isknown, then it is possible to estimate perceived video quality in a meaningfulway. However, automatic detection of viewers’ fixation points is time-consumingprocess and often is omitted in objective video quality assessment (VQA)metrics. Based on our previous work, in which we proposed Foveation-basedcontent Adaptive Root Mean Squared Error (FARMSE) VQA metric, in this work wepropose two new full-reference (FR) VQA metrics called Multi-Point FARMSE(MP-FARMSE), and Simple-FARMSE (S-FARMSE). Both new-proposed metrics are basedon foveated-vision features of human visual system and spatio-temporal featuresof video signal. In MP-FARMSE, by using an engineering approach, we implementedthe fact that viewer’s attention can be directed out of the center of theframe, thus covering use-cases when objects of interest are not located in thecenter of the frame. The main idea when creating the S-FARMSE metric was toreduce the computational complexity of the final algorithm and to make S-FARMSEmetric capable of processing high-resolution video signals in real-time.Performances of the new-proposed metrics are compared to performances of sevenexisting VQA metrics on two different video quality databases. The results showthat performances achieved by MP-FARMSE and S-FARMSE are quite close to thoseof state-of-the-art VQA metrics, whereas at the same time their computationalcomplexity level is significantly lower.
KW  - foveated vision
KW  - video quality
KW  - full reference
CR  - Glavota, Ivan; Kaprocki, Zvonimir; Vranješ, Mario; Herceg, Marijan., No-Reference Real-Time Video Transmission Artifacts Detection for Video Signals. // Journal of Real-Time Image Processing. 1 (2018) , 1; 1-22
CR  - Križanović, Višnja; Žagar, Drago; Grgić, Krešimir; Vranješ, Mario., Enhanced predictive modelling process of broadband services adoption based on time series data. // Advanced engineering informatics. 38 (2018) , October 2018; 142-167
CR  - Pul, Matija; Peković, Vukota; Vranješ, Mario; Grbić, Ratko., Automatic Functionality Verification of Hybrid Set-Top Boxes with Dynamic User Interface. // Ieee transactions on consumer electronics. 64 (2018) , 4; 1-9
CR  - Vranješ, Mario; Rimac-Drlje, Snježana; Vranješ, Denis., Foveation-based Content Adaptive Root Mean Squared Error for Video Quality Assessment. // Multimedia tools and applications. 77 (2018) , 16; 21053-21082
UR  - https://doi.org/10.22399/ijcesen.477034
L1  - https://dergipark.org.tr/tr/download/article-file/774647
ER  -