Stereoscopic Image Quality Assessment Network
Abstract
The goal of objective stereoscopic image quality assessment (SIQA) is to predict the human perceptual quality of stereoscopic/3D images automatically and accurately. Compared with traditional 2D image quality assessment (2D IQA), the quality assessment of stereoscopic images is more challenging because of the complex binocular vision mechanisms and multiple quality dimensions. In this paper, inspired by the hierarchical dual-stream interactive philosophy of the human visual system (HVS), we propose a Stereoscopic Image Quality Assessment Network (StereoQA-Net) for No-Reference stereoscopic image quality assessment (NR-SIQA). The proposed StereoQA-Net is an end-to-end dual-stream interactive network containing left and right view sub-networks, where the interaction of the two subnetworks presents in multiple layers. We evaluate our method on the LIVE stereoscopic image quality databases. Experimental results show that our proposed StereoQA-Net outperforms state-of-the-art algorithms on both symmetrically and asymmetrically distorted stereoscopic image pairs of various distortion types. And in more general case, the proposed StereoQA-Net can effectively predict the perceptual quality of local regions. In addition, cross-dataset experiments also demonstrate the generalization ability of our algorithm.
Paper: to be published.
Get Source Code
We develop a general-purpose NR-SIQA architecture, which takes the discriminative feature extraction and quality regression learning as an end-to-end optimization process.
To get the source code, Please fill THIS FORM and the source code will be sent to you.