Monocular-binocular feature fidelity induced index for stereoscopic image quality assessment

Perceptual three-dimensional image quality assessment (3D-IQA) aims to use computational models to measure image quality in a way that is consistent with human visual perception. Unlike previous studies that directly extended the two-dimensional metrics to the three-dimensional (3D) case, the major technical innovation of this paper is to simulate monocular and binocular visual perception and propose a monocular-binocular feature fidelity induced index for 3D-IQA. To be more specific, in the training stage, we train monocular and binocular dictionaries from the training database, so that the latent response properties can be represented as a set of basis vectors. In the quality estimation stage, we compute monocular feature fidelity and binocular feature fidelity indexes based on the estimated sparse coefficient vectors and compute a global energy response similarity index by considering global energy changes. The final quality score is obtained by incorporating them. Experimental results on four 3D-IQA databases demonstrate that in comparison with the most relevant existing methods, the devised algorithm achieves high consistency alignment with subjective assessment.