No-reference video quality assessment based on region of interest

In this paper, a novel no-reference video quality assessment algorithm based on region of interest (ROI) is introduced. Firstly, the video distortion is estimated by measuring the influence of blockiness and blur artifacts. Secondly, ROI is identified utilizing the encode information extracted from the bitstream with consideration of human vision system characteristics. Finally, the video quality is predicted using both video distortion and ROI. Experimental results show that the proposed algorithm can achieve better accuracy as compared with other methods.