Search papers, labs, and topics across Lattice.
This paper introduces UHD-GCN-BIQA, a graph representation learning framework for blind image quality assessment of ultra-high-definition (UHD) images. It addresses the challenge of balancing computational cost and the preservation of scale-sensitive distortions by sampling aspect-ratio-aligned patches and modeling their structural dependencies using a hybrid k-NN graph. The framework leverages residual graph convolution and gated attention pooling to achieve state-of-the-art performance on the UHD-IQA benchmark, demonstrating the effectiveness of graph-based region relation modeling.
Modeling relationships between image patches with graph neural networks substantially improves blind image quality assessment for ultra-high-definition images, achieving state-of-the-art results.
Blind image quality assessment (BIQA) for ultrahighdefinition (UHD) images remains challenging because native-resolution inference is computationally expensive, whereas aggressive resizing or isolated cropping may suppress scale-sensitive distortions and weaken the relationship between local artifacts and global scene context. This paper aims to improve UHD-BIQA by explicitly modeling the structural dependencies among sampled image regions rather than treating them as independent views, and a graph representation learning framework UHD-GCN-BIQA is proposed. The framework samples aspect-ratio-aligned patches from each UHD image, encodes them as graph nodes, and constructs a hybrid k-nearest-neighbor graph using spatial proximity and feature similarity. Residual graph convolution is used to propagate contextual information across regions, and gated attention pooling aggregates patchlevel evidence into an imagelevel quality prediction. An exponential moving average normalized multiobjective loss function is adopted to stabilize the joint optimization of regression, correlation, and ranking objectives. Experiments on the UHD-IQA benchmark show that UHD-GCN-BIQA achieves PLCC = 0.7784, SRCC = 0.8019, and RMSE = 0.0519, obtaining competitive correlation performance and the lowest RMSE among the compared methods. These results indicate that graph-based region relation modeling is effective for UHD image quality assessment, particularly for improving absolute quality score estimation under high-resolution visual content.
