AI RESEARCH

GraphLeap: Decoupling Graph Construction and Convolution for Vision GNN Acceleration on FPGA

arXiv CS.CV

ArXi:2604.21290v1 Announce Type: new Vision Graph Neural Networks (ViGs) represent an image as a graph of patch tokens, enabling adaptive, feature-driven neighborhoods. Unlike CNNs with fixed grid biases or Vision Transformers with global token interactions, ViGs rely on dynamic graph convolution: at each layer, a feature-dependent graph is built via k-nearest-neighbor (kNN) search on current patch features, followed by message passing.