Vision GNN: An Image is Worth Graph of Nodes

1 Jun 2022  ·  Kai Han, Yunhe Wang, Jianyuan Guo, Yehui Tang, Enhua Wu ·

Network architecture plays a key role in the deep learning-based computer vision system. The widely-used convolutional neural network and transformer treat the image as a grid or sequence structure, which is not flexible to capture irregular and complex objects. In this paper, we propose to represent the image as a graph structure and introduce a new Vision GNN (ViG) architecture to extract graph-level feature for visual tasks. We first split the image to a number of patches which are viewed as nodes, and construct a graph by connecting the nearest neighbors. Based on the graph representation of images, we build our ViG model to transform and exchange information among all the nodes. ViG consists of two basic modules: Grapher module with graph convolution for aggregating and updating graph information, and FFN module with two linear layers for node feature transformation. Both isotropic and pyramid architectures of ViG are built with different model sizes. Extensive experiments on image recognition and object detection tasks demonstrate the superiority of our ViG architecture. We hope this pioneering study of GNN on general visual tasks will provide useful inspiration and experience for future research. The PyTorch code is available at and the MindSpore code is available at

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Classification ImageNet Pyramid ViG-B Top 1 Accuracy 83.7% # 337
Number of params 92.6M # 813
GFLOPs 16.8 # 343
Image Classification ImageNet Pyramid ViG-S Top 1 Accuracy 82.1% # 491
Number of params 27.3M # 588
GFLOPs 4.6 # 211
Image Classification ImageNet Pyramid ViG-M Top 1 Accuracy 83.1% # 395
Number of params 51.7M # 696
GFLOPs 8.9 # 278
Image Classification ImageNet Pyramid ViG-Ti Top 1 Accuracy 78.2% # 736
Number of params 10.7M # 449
GFLOPs 1.7 # 132