In this paper we propose a reduced-reference quality assessment algorithm which computes an approximation of the Structural SIMilarity (SSIM) metrics exploiting coding tools provided by the distributed source coding theory. The algorithm has been tested to evaluate the quality of decoded video bitstreams after transmission over error-prone networks. We evaluate the accuracy of the proposed quality assessment algorithm by measuring the Pearson’s correlation coefficient between the structural similarity metrics computed in full-reference mode and the one provided by the proposed reduced-reference algorithm. The proposed reduced-reference algorithm achieves good correlation values (higher than 0.85 with packet loss rate equal up to 2.5%).
A reduced-reference structural similarity approximation for videos corrupted by channel errors
TAGLIASACCHI, MARCO;VALENZISE, GIUSEPPE;NACCARI, MATTEO;TUBARO, STEFANO
2010-01-01
Abstract
In this paper we propose a reduced-reference quality assessment algorithm which computes an approximation of the Structural SIMilarity (SSIM) metrics exploiting coding tools provided by the distributed source coding theory. The algorithm has been tested to evaluate the quality of decoded video bitstreams after transmission over error-prone networks. We evaluate the accuracy of the proposed quality assessment algorithm by measuring the Pearson’s correlation coefficient between the structural similarity metrics computed in full-reference mode and the one provided by the proposed reduced-reference algorithm. The proposed reduced-reference algorithm achieves good correlation values (higher than 0.85 with packet loss rate equal up to 2.5%).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
