Objective Quality Assessment Method for Stereoscopic Image Retargeting

Abstract

The popularity and variety of stereoscopic display devices have led to the need for stereoscopic image retargeting (SIR). Evaluating the quality of retargeted stereoscopic images objectively has become of great importance for optimizing the retargeting operation. The previous works have leveraged 2D image retargeting quality assessment methods to address this issue. However, these methods cannot effectively detect image structure distortion, which is a crucial factor for quality degradation of retargeted stereoscopic images. In this paper, we first propose a novel triangulation-based method to effectively represent the structure distortion and image quality. And then we further evaluate the retargeted stereoscopic images from the perspective of image quality, depth perception and visual comfort. Experimental results demonstrate that our method outperforms the state-of-the-art methods in the consistency level of the predicted scores with subjective scores.