Objective quality assessment for compressed video
Susu Yao, Weisi Lin, Zhongkang Lu, Ee Ping Ong, Minoru Etoh · 2003
We propose a new objective quality metric for evaluating image and video quality in coding applications. In this metric, visual quality scores are given based on a combination of three factors: visually masked error, blurring distortion and structural distortion. Compared with vision-based models, it is easier to calculate and applicable to both image and video processing applications. Good correlation with subjective test scores has been obtained using the proposed metric.