HSNet: Heterogeneous Subgraph Network for Single Image Super-resolution
Qiongyang Hu, Wenyang Liu, Wenbin Zou, Yuejiao Su, Lap-Pui Chau, Yi Wang
TL;DR
HSNet tackles the rigidity and high cost of CNN- and attention-based SR by introducing a graph-based approach that represents image features through a heterogeneous set of subgraphs. It combines Construct Subgraph Set Block (CSSB) for diverse subgraph generation, Subgraph Aggregation Block (SAB) for multi-view fusion via Graph Aggregation and Learnable Parameters, and a Node Sampling Strategy (NSS) to reduce computation while preserving salient information. Across five standard SR benchmarks, HSNet achieves state-of-the-art reconstruction quality with lower parameters and FLOPs than many baselines, validating its effectiveness and efficiency. The results indicate that multi-subgraph perspective fusion yields robust texture preservation and details, with practical potential for real-world SR systems and public code release.
Abstract
Existing deep learning approaches for image super-resolution, particularly those based on CNNs and attention mechanisms, often suffer from structural inflexibility. Although graph-based methods offer greater representational adaptability, they are frequently impeded by excessive computational complexity. To overcome these limitations, this paper proposes the Heterogeneous Subgraph Network (HSNet), a novel framework that efficiently leverages graph modeling while maintaining computational feasibility. The core idea of HSNet is to decompose the global graph into manageable sub-components. First, we introduce the Constructive Subgraph Set Block (CSSB), which generates a diverse set of complementary subgraphs. Rather than relying on a single monolithic graph, CSSB captures heterogeneous characteristics of the image by modeling different relational patterns and feature interactions, producing a rich ensemble of both local and global graph structures. Subsequently, the Subgraph Aggregation Block (SAB) integrates the representations embedded across these subgraphs. Through adaptive weighting and fusion of multi-graph features, SAB constructs a comprehensive and discriminative representation that captures intricate interdependencies. Furthermore, a Node Sampling Strategy (NSS) is designed to selectively retain the most salient features, thereby enhancing accuracy while reducing computational overhead. Extensive experiments demonstrate that HSNet achieves state-of-the-art performance, effectively balancing reconstruction quality with computational efficiency. The code will be made publicly available.