DVHGNN: Multi-Scale Dilated Vision HGNN for Efficient Vision Recognition
This work addresses efficiency and relational limitations in vision recognition for computer vision researchers, offering incremental improvements over existing graph-based methods.
The paper tackles the quadratic computational complexity and limited pairwise relations in Vision Graph Neural Networks by proposing DVHGNN, which uses multi-scale hypergraphs to capture high-order correlations, achieving a top-1 accuracy of 83.1% on ImageNet-1K, surpassing prior models by up to +1.0%.
Recently, Vision Graph Neural Network (ViG) has gained considerable attention in computer vision. Despite its groundbreaking innovation, Vision Graph Neural Network encounters key issues including the quadratic computational complexity caused by its K-Nearest Neighbor (KNN) graph construction and the limitation of pairwise relations of normal graphs. To address the aforementioned challenges, we propose a novel vision architecture, termed Dilated Vision HyperGraph Neural Network (DVHGNN), which is designed to leverage multi-scale hypergraph to efficiently capture high-order correlations among objects. Specifically, the proposed method tailors Clustering and Dilated HyperGraph Construction (DHGC) to adaptively capture multi-scale dependencies among the data samples. Furthermore, a dynamic hypergraph convolution mechanism is proposed to facilitate adaptive feature exchange and fusion at the hypergraph level. Extensive qualitative and quantitative evaluations of the benchmark image datasets demonstrate that the proposed DVHGNN significantly outperforms the state-of-the-art vision backbones. For instance, our DVHGNN-S achieves an impressive top-1 accuracy of 83.1% on ImageNet-1K, surpassing ViG-S by +1.0% and ViHGNN-S by +0.6%.