Quality improvement of H.264/AVC FRExt by incorporating perceptual models

Abstract

This paper presents the quality improvement of the current video compression standard by incorporating Human Vision Models (HVS) into the standard coding structure. The advanced video coding (AVC) which is also called as H.264/MPEG-4 is the latest compression standard available for video compression. Fidelity Range Extension (FRExt) is the recent work done on the AVC which includes a number of enhanced capabilities relative to the base standards. FRExt aims at improving the quality of the reconstructed videos for some professional applications which require high resolution, such as video conferencing, video broadcast and studio quality videos. In this work a new codec incorporating HVS model has been proposed which gives improved visual quality compared to the MPEG 2, MPEG-4 AVC and FRExt. The proposed coder uses the Contrast Sensitivity Function (CSF) in the transform coding stage of the standard codec. To see the performance improvements of the proposed coder, AVC with FRExt features has been implemented first and the quality of the reconstructed video signals are evaluated using both subjective and objective measures such as MOS, PSNR, MSSIM, VIF and VSNR. The proposed coder shows a noticeable improvement in perceptual quality compared to the standard coders which promises scope for further improvements in the area of perceptual based video coders.