Quality Assessment for Stereoscopic Images by Distortion Separation

Quality assessment for stereoscopic images is a challenging issue in three-dimensional research. In this paper, we present an objective quality assessment method for stereoscopic images by distortion separation. In the method, we separate the distortion information for the distorted stereoscopic image (i.e., decompose the distorted stereoscopic image into restored and disturbed stereoscopic images), and use phase-amplitude description model and singular value decomposition model to evaluate them respectively. The experimental results show that compared with other schemes, the proposed method can achieve much higher consistency with the subjective assessments.