Blind video quality assessment based on spatio-temporal internal generative mechanism | IEEE Conference Publication | IEEE Xplore

Blind video quality assessment based on spatio-temporal internal generative mechanism


Abstract:

In this paper, we present a new blind video quality assessment metric considering the characteristics of human visual system (HVS). HVS indicates that internal generative...Show More

Abstract:

In this paper, we present a new blind video quality assessment metric considering the characteristics of human visual system (HVS). HVS indicates that internal generative mechanism (IGM) actively predicts the visual scene (predicted information) and avoid the residual uncertainty (uncertain information). Based on spatio-temporal internal generative mechanism (ST-IGM), we employ a spatio-temporal autoregressive (AR) prediction model to disassemble the video content into the predicted part and the uncertain part. Then, we separately evaluate their quality degradations by natural video statistics (NVS) model based blind VQA method. According to the perception of distortions in two parts, different weights are assigned to yield the overall quality. The experimental results demonstrate that the proposed algorithm performs much better than the state-of-the-art blind train-free algorithm on the LIVE VQA database and shows competitive performance with the existing train-based methods.
Date of Conference: 17-20 September 2017
Date Added to IEEE Xplore: 22 February 2018
ISBN Information:
Electronic ISSN: 2381-8549
Conference Location: Beijing, China

References

References is not available for this document.