(A) study on no-reference video quality assessment using block modes, motion vectors, boundary strength and quantization parameters of h.264/avc

Abstract

Since demands for video services have been rapidly increasing, the require-ments of video quality assessment are also increasing. Although subjective quali-ty assessment is more reliable due to the reflection of Human Visual System (HVS), it is a time-consuming and very expensive approach. Also it is not useful for real-time applications. Therefore, much research has been made for objective video quality assessment instead of subjective video quality assessment. Among three kinds of objective assessment which are full-reference, re-duced-reference and no-reference methods, no-reference method has drawn much attention because it does not require any reference. The encoding parame-ters are good features to use for no-reference model in video quality assessment because the encoded bitstreams carry plenty of information about the video con-tents and it is easy to extract some coding parameters without extra processing in pixel domain. Some previous quality metrics which has utilized coding parame-ters have been studied. In this thesis, we propose a no-reference quality metric using various kinds of coding parameters. We use the quantization parameters to measure the overall visual quality of video and motion vectors to estimate temporal characteristic of video. Block mode parameters are extracted to calculate spatial characteristic and boundary strength information is used to estimate sharpness or blurring degree of reconstructed video contents. All of these parameters are extracted and computed from H.264|AVC bitstreams, without relying on pixel domain processing. We design a linear quality metric composed of these four parameters for video quality assessment model. The model parameters are estimated using linear regression with the results of subjective quality assessment which are ob-tained by conducting the DSIS (Double Stimulus Impairment Scale) method of ITU- R BT.500-11. The experimental result shows that, based on the proposed quality metric, the propos...