mlsanityData to 7 Oct 2026

Restore-RWKV: Efficient and Effective Medical Image Restoration with RWKV

arXiv 2407.11087 ↗First posted 14 Jul 2024eess.IV cs.CVcited by 36

Not on the map. Why?

Abstract (arXiv)

Transformers have revolutionized medical image restoration, but the quadratic complexity still poses limitations for their application to high-resolution medical images. The recent advent of the Receptance Weighted Key Value (RWKV) model in the natural language processing field has attracted much attention due to its ability to process long sequences efficiently. To leverage its advanced design, we propose Restore-RWKV, the first RWKV-based model for medical image restoration. Since the original RWKV model is designed for 1D sequences, we make two necessary modifications for modeling spatial relations in 2D medical images. First, we present a recurrent WKV (Re-WKV) attention mechanism that captures global dependencies with linear computational complexity. Re-WKV incorporates bidirectional attention as basic for a global receptive field and recurrent attention to effectively model 2D dependencies from various scan directions. Second, we develop an omnidirectional token shift (Omni-Shift) layer that enhances local dependencies by shifting tokens from all directions and across a wide context range. These adaptations make the proposed Restore-RWKV an efficient and effective model for medical image restoration. Even a lightweight variant of Restore-RWKV, with only 1.16 million parameters, achieves comparable or even superior results compared to existing state-of-the-art (SOTA) methods. Extensive experiments demonstrate that the resulting Restore-RWKV achieves SOTA performance across a range of medical image restoration tasks, including PET image synthesis, CT image denoising, MRI image super-resolution, and all-in-one medical image restoration. Code is available at: https://github.com/Yaziwel/Restore-RWKV.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Builds on 47 of its 70 references that are in the map. First posted on arXiv 14 Jul 2024.

Builds on

47 of 70 references.

References · 47

  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. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  5. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  6. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  7. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,869
  8. 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,618
  9. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  10. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,218
  11. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  12. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
35 more
  1. 2024VMamba: Visual State Space Modelcited by 955
  2. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  3. 2023RWKV: Reinventing RNNs for the Transformer Eracited by 814
  4. 2017Understanding the Effective Receptive Field in Deep Convolutional Neural Networkscited by 549
  5. 2021RepVGG: Making VGG-style ConvNets Great Againcited by 520
  6. 2022MaxViT: Multi-Axis Vision Transformercited by 476
  7. 2024MambaIR: A Simple Baseline for Image Restoration with State-Space Modelcited by 349
  8. 2017Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)cited by 224
  9. 2017Burst Denoising with Kernel Prediction Networkscited by 156
  10. 2024Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrencecited by 155
  11. 2024Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architecturescited by 97
  12. 2018Efficient and Accurate MRI Super-Resolution using a Generative Adversarial Network and 3D Multi-Level Densely Connected Networkcited by 64
  13. 2023SparseViT: Revisiting Activation Sparsity for Efficient High-Resolution Vision Transformercited by 60
  14. 2022Swin Transformer for Fast MRIcited by 56
  15. 2024Towards Effective Multiple-in-One Image Restoration: A Sequential and Prompt Learning Strategycited by 48
  16. 2022CTformer: Convolution-free Token2Token Dilated Vision Transformer for Low-dose CT Denoisingcited by 46
  17. 2024Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Modelscited by 41
  18. 2024Visual Mamba: A Survey and New Outlookscited by 41
  19. 2023Xformer: Hybrid X-Shaped Transformer for Image Denoisingcited by 37
  20. 2021TransCT: Dual-path Transformer for Low Dose Computed Tomographycited by 35
  21. 2021Eformer: Edge Enhancement based Transformer for Medical Image Denoisingcited by 30
  22. 2024All-In-One Medical Image Restoration via Task-Adaptive Routingcited by 30
  23. 2018Channel Splitting Network for Single MR Image Super-Resolutioncited by 26
  24. 2020EDCNN: Edge enhancement-based Densely Connected Network with Compound Loss for Low-Dose CT Denoisingcited by 26
  25. 2024RWKV-CLIP: A Robust Vision-Language Representation Learnercited by 23
  26. 2024PointRWKV: Efficient RWKV-Like Model for Hierarchical Point Cloud Learningcited by 21
  27. 2022Spach Transformer: Spatial and Channel-wise Transformer Based on Local and Global Self-attentions for PET Image Denoisingcited by 19
  28. 2017200x Low-dose PET Reconstruction using Deep Learningcited by 18
  29. 2024Mamba or RWKV: Exploring High-Quality and High-Efficiency Segment Anything Modelcited by 18
  30. 2024Deform-Mamba Network for MRI Super-Resolutioncited by 17
  31. 2022Swin Deformable Attention U-Net Transformer (SDAUT) for Explainable Fast MRIcited by 14
  32. 20223D Segmentation Guided Style-based Generative Adversarial Networks for PET Synthesiscited by 13
  33. 2023DRMC: A Generalist Model with Dynamic Routing for Multi-Center PET Image Synthesiscited by 8
  34. 2024IPT-V2: Efficient Image Processing Transformer using Hierarchical Attentionscited by 7
  35. 2024No-Clean-Reference Image Super-Resolution: Application to Electron Microscopycited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 18; 2026: 13.

36 papers.

Sort:

By area:

  1. 2025Linear Attention Modeling for Learned Image Compressioncited by 22
  2. 2025NTIRE 2025 Challenge on Image Super-Resolution (x4): Methods and Resultscited by 12
  3. 2025RWKV-UNet: Improving UNet with Long-Range Cooperation for Effective Medical Image Segmentationcited by 7
  4. 2025MRI Super-Resolution with Deep Learning: A Comprehensive Surveycited by 7
  5. 2026Learning Domain-Aware Task Prompt Representations for Multi-Domain All-in-One Image Restorationcited by 6
  6. 2024A Survey of RWKVcited by 5
  7. 2025C3-OWD: A Curriculum Cross-modal Contrastive Learning Framework for Open-World Detectioncited by 5
  8. 2024Adaptive Blind All-in-One Image Restorationcited by 3
  9. 2025FEAT: Full-Dimensional Efficient Attention Transformer for Medical Video Generationcited by 3
  10. 2025TAT: Task-Adaptive Transformer for All-in-One Medical Image Restorationcited by 3
  11. 2024StyleRWKV: High-Quality and High-Efficiency Style Transfer with RWKV-like Architecturecited by 2
  12. 2025Multi-View Learning with Context-Guided Receptance for Image Denoisingcited by 2
24 more
  1. 2025EvRWKV: A Continuous Interactive RWKV Framework for Effective Event-Guided Low-Light Image Enhancementcited by 2
  2. 2025EndoIR: Degradation-Agnostic All-in-One Endoscopic Image Restoration via Noise-Aware Routing Diffusioncited by 2
  3. 2024The Evolution of RWKV: Advancements in Efficient Language Modelingcited by 1
  4. 2025DREMnet: An Interpretable Denoising Framework for Semi-Airborne Transient Electromagnetic Signalcited by 1
  5. 2025UniFuse: A Unified All-in-One Framework for Multi-Modal Medical Image Fusion Under Diverse Degradations and Misalignmentscited by 1
  6. 2025FS-RWKV: Leveraging Frequency Spatial-Aware RWKV for 3T-to-7T MRI Translationcited by 1
  7. 2025MRT: Learning Compact Representations with Mixed RWKV-Transformer for Extreme Image Compressioncited by 1
  8. 2025Fourier-RWKV: A Multi-State Perception Network for Efficient Image Dehazingcited by 1
  9. 2026Towards Controllable Low-Light Image Enhancement: A Continuous Multi-illumination Dataset and Efficient State Space Frameworkcited by 1
  10. 2024GDSR: Global-Detail Integration through Dual-Branch Network with Wavelet Losses for Remote Sensing Image Super-Resolutioncited by 0
  11. 2025Exploring Linear Attention Alternative for Single Image Super-Resolutioncited by 0
  12. 2025RSRWKV: A Linear-Complexity 2D Attention Mechanism for Efficient Remote Sensing Vision Taskcited by 0
  13. 2025Exploring Diffusion with Test-Time Training on Efficient Image Restorationcited by 0
  14. 2026Revisiting Global Token Mixing in Task-Dependent MRI Restoration: Insights from Minimal Gated CNN Baselinescited by 0
  15. 2026MedQ-UNI: Toward Unified Medical Image Quality Assessment and Restoration via Vision-Language Modelingcited by 0
  16. 2026EventFace: Event-Based Face Recognition via Structure-Driven Spatiotemporal Modelingcited by 0
  17. 2026PestVL-Net: Enabling Multimodal Pest Learning via Fine-grained Vision-Language Interactioncited by 0
  18. 2026SCRWKV: Ultra-Compact Structure-Calibrated Vision-RWKV for Topological Crack Segmentationcited by 0
  19. 2026U-TTT: Towards Generalizable PET Image Denoising via Test-Time Trainingcited by 0
  20. 2026CLUIE: Clustering-Aware Recurrent Propagation with Local Structural Compensation for Underwater Image Enhancementcited by 0
  21. 2026Compass: Degradation-Simulated Reciprocal Learning with Lightweight Needle RWKV for Multimodal Crack Segmentation under Missing Modalitiescited by 0
  22. 2026NeuroPath: Brain-Inspired Dual-Pathway Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 0
  23. 2026MARR: Decoupling Policy, Execution, and Calibration for All-in-One Medical Image Restorationcited by 0
  24. 2026UniH3: Unifying Hierarchical Homogeneity and Heterogeneity for All-in-One Medical Image Restorationcited by 0