Saved in:
Bibliographic Details
Main Authors: Yang, Hao, Zhang, Xu, Ma, Jiaqi, Zhu, Linwei, Zhang, Yun, Zhang, Huan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.09843
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909735897268224
author Yang, Hao
Zhang, Xu
Ma, Jiaqi
Zhu, Linwei
Zhang, Yun
Zhang, Huan
author_facet Yang, Hao
Zhang, Xu
Ma, Jiaqi
Zhu, Linwei
Zhang, Yun
Zhang, Huan
contents Current Omnidirectional Image Quality Assessment (OIQA) methods struggle to evaluate locally non-uniform distortions due to inadequate modeling of spatial variations in quality and ineffective feature representation capturing both local details and global context. To address this, we propose a graph neural network-based OIQA framework that explicitly models structural relationships between viewports to enhance perception of spatial distortion non-uniformity. Our approach employs Fibonacci sphere sampling to generate viewports with well-structured topology, representing each as a graph node. Multi-stage feature extraction networks then derive high-dimensional node representation. To holistically capture spatial dependencies, we integrate a Graph Attention Network (GAT) modeling fine-grained local distortion variations among adjacent viewports, and a graph transformer capturing long-range quality interactions across distant regions. Extensive experiments on two large-scale OIQA databases with complex spatial distortions demonstrate that our method significantly outperforms existing approaches, confirming its effectiveness and strong generalization capability.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09843
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Graph Attention Network for No-Reference Omnidirectional Image Quality Assessment
Yang, Hao
Zhang, Xu
Ma, Jiaqi
Zhu, Linwei
Zhang, Yun
Zhang, Huan
Computer Vision and Pattern Recognition
Current Omnidirectional Image Quality Assessment (OIQA) methods struggle to evaluate locally non-uniform distortions due to inadequate modeling of spatial variations in quality and ineffective feature representation capturing both local details and global context. To address this, we propose a graph neural network-based OIQA framework that explicitly models structural relationships between viewports to enhance perception of spatial distortion non-uniformity. Our approach employs Fibonacci sphere sampling to generate viewports with well-structured topology, representing each as a graph node. Multi-stage feature extraction networks then derive high-dimensional node representation. To holistically capture spatial dependencies, we integrate a Graph Attention Network (GAT) modeling fine-grained local distortion variations among adjacent viewports, and a graph transformer capturing long-range quality interactions across distant regions. Extensive experiments on two large-scale OIQA databases with complex spatial distortions demonstrate that our method significantly outperforms existing approaches, confirming its effectiveness and strong generalization capability.
title Hierarchical Graph Attention Network for No-Reference Omnidirectional Image Quality Assessment
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.09843