Abstract:The emergence of virtual reality technology has made stereoscopic omnidirectional images (SOI) easily accessible and prompted the need to evaluate their perceptual quality. At present, most stereoscopic omnidirectional image quality assessment (SOIQA) methods rely on one of the projection formats, i.e., Equirectangular Projection (ERP) or CubeMap Projection (CMP). However, while ERP provides global information and the less distorted CMP complements it by providing local structural guidance, research on leveraging both ERP and CMP in SOIQA remains limited, hindering a comprehensive understanding of both global and local visual cues. Motivated by this gap, our study introduces a novel dual-stream perception-driven network for blind quality assessment of stereoscopic omnidirectional images. By integrating both ERP and CMP, our method effectively captures both global and local information, marking the first attempt to bridge this gap in SOIQA, particularly through deep learning methodologies. We employ an inter-intra feature fusion module, which considers both the inter-complementarity between ERP and CMP and the intra-relationships within CMP images. This module dynamically and complementarily adjusts the contributions of features from both projections and effectively integrates them to achieve a more comprehensive perception. Besides, deformable convolution is employed to extract the local region of interest, simulating the orientation selectivity of the primary visual cortex. Finally, with the features of left and right views of SOI, a stereo cross attention module that simulates the binocular fusion mechanism is proposed to predict the quality score. Extensive experiments are conducted to evaluate our model and the state-of-the-art competitors, demonstrating that our model has achieved the best performance on the databases of LIVE 3D VR, SOLID, and NBU.

Subjective Quality Assessment of Stereoscopic Omnidirectional Image.

Stereoscopic Omnidirectional Image Quality Assessment Based on Predictive Coding Theory

Assessment of Perceptual Depth Intensity on Source Stereoscopic Images

Quality Assessment of Stereoscopic 360-degree Images from Multi-viewports

Perceptual Depth Quality Assessment of Stereoscopic Omnidirectional Images

Viewport Perception Based Blind Stereoscopic Omnidirectional Image Quality Assessment

Quality of experience assessment for stereoscopic images.

Projection Invariant Feature and Visual Saliency-Based Stereoscopic Omnidirectional Image Quality Assessment

Subjective Perception Assessment of Stereoscopic Image Quality：Study

Assessment Of 3d Quality Of Experience With Content Analysis

Stereoscopic Image Quality Assessment Based on Depth and Texture Information

On Evaluation the Quality of Subjective S3D Comfort Assessment

Research on Subjective Stereoscopic Image Quality Assessment

Dual-stream Perception-driven Blind Quality Assessment for Stereoscopic Omnidirectional Images

Perceptual Quality Assessment of Omnidirectional Images

Stereoscopic Image Quality Assessment Combining Statistical Features and Binocular Theory

Models of Monocular and Binocular Visual Perception in Quality Assessment of Stereoscopic Images

Monocular and Binocular Interactions Oriented Deformable Convolutional Networks for Blind Quality Assessment of Stereoscopic Omnidirectional Images

A New Objective Stereoscopic Image Assessment Model Based on Stereoscopic Perception

Perceptual Quality Assessment of Omnidirectional Images: A Benchmark and Computational Model

Perceptual Quality Assessment of Omnidirectional Images:Subjective Experiment and Objective Model Evaluation