No-reference image quality assessment based on BNB measurement

Fang, Ruigang; Wu, Dapeng

doi:10.1109/chinasip.2013.6625396

Cited by 5 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…BNB metrics are proposed in our previous work [9] , which could measure image blurriness, noisiness and blockiness well. In this paper, we employ these three features and (1) Codebook Construction The ith element of the codebook is a vector including five values: blurriness feature, noisiness feature, blockiness feature, viewing distance, QoE score, which are noted as 5 .…”

Section: Image Qoe Predictionmentioning

confidence: 99%

“…Considerable attention has been attracted to the research on image quality assessment, such as Full Reference Image Quality Assessment (FRIQA) [1] [2] [3], Reduced Reference Image Quality Assessment (RRIQA) [4] [5] [6], and No Reference Image Quality Assessment [7] [8] [9]. All these above methods only consider image itself, and they assume that the images are viewed under the same condition for all users.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of image quality of experience in consideration of viewing distance

Fang

Shen

2015

2015 IEEE China Summit and International Conference on Signal and Information Processing (ChinaSIP)

Self Cite

View full text Add to dashboard Cite

Image viewing distance plays an important role in the assessment of image Quality of Experience (QoE). In this work, we present a subjective image QoE study in which a total of 494 images evaluated by more than 30 human subjects at 7 different viewing distance. Through the study, we observed that different images have different regularities between viewing distance and their QoE. A No-Reference QoE assessment model is proposed to objectively measure image QoE considering viewing distance. The experiments conducted on our database show that the proposed model achieves high correlation between its predicted QoE score and human perception. Moreover, we have made the image database freely available to the research community.

show abstract

Section: Image Qoe Predictionmentioning

confidence: 99%