mlsanityData to 7 Oct 2026

Vision GNN: An Image is Worth Graph of Nodes

arXiv 2206.00272 ↗First posted 1 Jun 2022cs.CVcited by 185

Not on the map. Why?

Abstract (arXiv)

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 https://github.com/huawei-noah/Efficient-AI-Backbones and the MindSpore code is available at https://gitee.com/mindspore/models.

Key facts. Cited by 185 papers in the map (data through 7 Oct 2026). Builds on 62 of its 86 references that are in the map. First posted on arXiv 1 Jun 2022.

Builds on

62 of 86 references.

References · 62

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  6. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  7. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  8. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  9. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  10. 2020End-to-End Object Detection with Transformerscited by 7,384
  11. 2017Mask R-CNNcited by 7,259
  12. 2017Inductive Representation Learning on Large Graphscited by 6,284
38 more
  1. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  2. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  3. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  4. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  5. 2018How Powerful are Graph Neural Networks?cited by 4,355
  6. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  7. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  8. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  9. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  10. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  11. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  12. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  13. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
  14. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,960
  15. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  16. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
  17. 2017Random Erasing Data Augmentationcited by 1,235
  18. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
  19. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  20. 2013Spectral Networks and Locally Connected Networks on Graphscited by 1,097
  21. 2016Deep Networks with Stochastic Depthcited by 973
  22. 2021CvT: Introducing Convolutions to Vision Transformerscited by 946
  23. 2020Pre-Trained Image Processing Transformercited by 891
  24. 2020A Survey on Visual Transformercited by 653
  25. 2021Transformer in Transformercited by 632
  26. 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
  27. 2021MetaFormer Is Actually What You Need for Visioncited by 523
  28. 2017Scene Graph Generation by Iterative Message Passingcited by 462
  29. 2018Lipschitz regularity of deep neural networks: analysis and efficient estimationcited by 413
  30. 2021Attention is Not All You Need: Pure Attention Loses Rank Doubly Exponentially with Depthcited by 411
  31. 2017Large-scale Point Cloud Semantic Segmentation with Superpoint Graphscited by 384
  32. 2021ResNet strikes back: An improved training procedure in timmcited by 365
  33. 2021Bottleneck Transformers for Visual Recognitioncited by 364
  34. 2021ResMLP: Feedforward networks for image classification with data-efficient trainingcited by 338
  35. 2016Learning Convolutional Neural Networks for Graphscited by 314
  36. 2018Graph R-CNN for Scene Graph Generationcited by 296
  37. 2015Diffusion-Convolutional Neural Networkscited by 268
  38. 2015Deep Convolutional Networks on Graph-Structured Datacited by 246

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 14; 2023: 47; 2024: 44; 2025: 53; 2026: 27.

185 papers.

Sort:

By area:

  1. 2022MetaFormer Baselines for Visioncited by 131
  2. 2022A Generalization of ViT/MLP-Mixer to Graphscited by 93
  3. 2023Pushing the Limits of Fewshot Anomaly Detection in Industry Vision: Graphcorecited by 75
  4. 2024GROUNDHOG: Grounding Large Language Models to Holistic Segmentationcited by 70
  5. 2022On the Trade-off between Over-smoothing and Over-squashing in Deep Graph Neural Networkscited by 63
  6. 2022A Survey on Graph Neural Networks and Graph Transformers in Computer Vision: A Task-Oriented Perspectivecited by 56
  7. 2022Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformerscited by 47
  8. 2023Image as Set of Pointscited by 43
  9. 2023MobileViG: Graph-Based Sparse Attention for Mobile Vision Applicationscited by 35
  10. 2024Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation?cited by 31
  11. 2023Masked Contrastive Graph Representation Learning for Age Estimationcited by 27
  12. 2023GazeGNN: A Gaze-Guided Graph Neural Network for Chest X-ray Classificationcited by 25
38 more
  1. 2023Graph Information Bottleneck for Remote Sensing Segmentationcited by 24
  2. 2023FFT-based Dynamic Token Mixer for Visioncited by 22
  3. 2023GraphEcho: Graph-Driven Unsupervised Domain Adaptation for Echocardiogram Video Segmentationcited by 22
  4. 2023G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentationcited by 21
  5. 2024GreedyViG: Dynamic Axial Graph Construction for Efficient Vision GNNscited by 21
  6. 2024Gramformer: Learning Crowd Counting via Graph-Modulated Transformercited by 19
  7. 2025Graph-based Approaches and Functionalities in Retrieval-Augmented Generation: A Comprehensive Surveycited by 18
  8. 2023Granular-ball computing: an efficient, robust, and interpretable adaptive multi-granularity representation and computation methodcited by 16
  9. 2023A Graph-Based Approach for Category-Agnostic Pose Estimationcited by 16
  10. 2025Hypergraph Vision Transformers: Images are More than Nodes, More than Edgescited by 16
  11. 2023Multi-scale Promoted Self-adjusting Correlation Learning for Facial Action Unit Detectioncited by 14
  12. 2023Computation-efficient Deep Learning for Computer Vision: A Surveycited by 14
  13. 2024Sharing Key Semantics in Transformer Makes Efficient Image Restorationcited by 14
  14. 2024GRPose: Learning Graph Relations for Human Image Generation with Pose Priorscited by 14
  15. 2023RIFormer: Keep Your Vision Backbone Effective While Removing Token Mixercited by 13
  16. 2024TCFormer: Visual Recognition via Token Clustering Transformercited by 12
  17. 2024Continual Learning on Graphs: A Surveycited by 11
  18. 2026OpenMAG: A Comprehensive Benchmark for Multimodal-Attributed Graphcited by 11
  19. 2022Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactionscited by 10
  20. 2023SLSG: Industrial Image Anomaly Detection by Learning Better Feature Embeddings and One-Class Classificationcited by 10
  21. 2023PVG: Progressive Vision Graph for Vision Recognitioncited by 10
  22. 2024Explaining Graph Neural Networks via Structure-aware Interaction Indexcited by 10
  23. 2022Semantic2Graph: Graph-based Multi-modal Feature Fusion for Action Segmentation in Videoscited by 9
  24. 2023NexToU: Efficient Topology-Aware U-Net for Medical Image Segmentationcited by 9
  25. 2023MaGNAS: A Mapping-Aware Graph Neural Architecture Search Framework for Heterogeneous MPSoC Deploymentcited by 9
  26. 2024Scaling Graph Convolutions for Mobile Visioncited by 9
  27. 2024WiGNet: Windowed Vision Graph Neural Networkcited by 8
  28. 2023ParameterNet: Parameters Are All You Needcited by 7
  29. 2023Keypoint-Augmented Self-Supervised Learning for Medical Image Segmentation with Limited Annotationcited by 7
  30. 2023Semantic-Aware Frame-Event Fusion based Pattern Recognition via Large Vision-Language Modelscited by 7
  31. 2024Gaze-directed Vision GNN for Mitigating Shortcut Learning in Medical Imagecited by 7
  32. 2024A Multi-view Mask Contrastive Learning Graph Convolutional Neural Network for Age Estimationcited by 7
  33. 2023Unsupervised HDR Image and Video Tone Mapping via Contrastive Learningcited by 6
  34. 2023ClusterFormer: Clustering As A Universal Visual Learnercited by 6
  35. 2025ClusterViG: Efficient Globally Aware Vision GNNs via Image Partitioningcited by 6
  36. 2025DVHGNN: Multi-Scale Dilated Vision HGNN for Efficient Vision Recognitioncited by 6
  37. 2025TopoLiDM: Topology-Aware LiDAR Diffusion Models for Interpretable and Realistic LiDAR Point Cloud Generationcited by 6
  38. 2023VSViG: Real-time Video-based Seizure Detection via Skeleton-based Spatiotemporal ViGcited by 5

50 most cited of 185 All 185 (JSON)