An Efficient Reduced-reference Video Quality Metric

Abstract : We present an efficient reduced−reference video quality assessment method based on local harmonic strength (LHS) feature. LHS is based on harmonics gain and loss information extracted from picture that correspond with the two most prominent compression artifacts, blockiness and blurriness. Efficiency is achieved through further reduction of the reduced−reference overhead. For the LHS based method, we used quantization of the LHS feature, temporal sampling, and block selection with segmentation−based block classification method. A typical reduced−reference low overhead data rate (around 0.3−1 kbps) with good prediction performance (0.82− 0.87 Pearson correlations) for broadcast type video sequences in VQEG Test Phase−I datasets has been achieved by the proposed method. With such low bit rate overhead data, the method proposed in this paper can be useful for continuous monitoring of video quality applications.