mlsanityData to 7 Oct 2026

MambaIR: A Simple Baseline for Image Restoration with State-Space Model

arXiv 2402.15648 ↗First posted 23 Feb 2024cs.CVcited by 349

On the map: State-space sequence models (S4, Mamba) → Mamba for image restoration

Abstract (arXiv)

Recent years have seen significant advancements in image restoration, largely attributed to the development of modern deep neural networks, such as CNNs and Transformers. However, existing restoration backbones often face the dilemma between global receptive fields and efficient computation, hindering their application in practice. Recently, the Selective Structured State Space Model, especially the improved version Mamba, has shown great potential for long-range dependency modeling with linear complexity, which offers a way to resolve the above dilemma. However, the standard Mamba still faces certain challenges in low-level vision such as local pixel forgetting and channel redundancy. In this work, we introduce a simple but effective baseline, named MambaIR, which introduces both local enhancement and channel attention to improve the vanilla Mamba. In this way, our MambaIR takes advantage of the local pixel similarity and reduces the channel redundancy. Extensive experiments demonstrate the superiority of our method, for example, MambaIR outperforms SwinIR by up to 0.45dB on image SR, using similar computational cost but with a global receptive field. Code is available at urlhttps://github.com/csguoh/MambaIR.

Key facts. Cited by 349 papers in the map (data through 7 Oct 2026). Placed in Mamba for image restoration, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 63 of its 96 references that are in the map. First posted on arXiv 23 Feb 2024.

Builds on

63 of 96 references.

References · 63

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  5. 2020End-to-End Object Detection with Transformerscited by 7,384
  6. 2017Squeeze-and-Excitation Networkscited by 4,763
  7. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  8. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  9. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  10. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,869
  11. 2020GLU Variants Improve Transformercited by 1,724
  12. 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,618
38 more
  1. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  2. 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,352
  3. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,256
  4. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
  5. 2021Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Datacited by 961
  6. 2024VMamba: Visual State Space Modelcited by 955
  7. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  8. 2020Pre-Trained Image Processing Transformercited by 891
  9. 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 864
  10. 2021Multi-Stage Progressive Image Restorationcited by 813
  11. 2018Residual Dense Network for Image Super-Resolutioncited by 796
  12. 2022Simplified State Space Layers for Sequence Modelingcited by 776
  13. 2022Simple Baselines for Image Restorationcited by 701
  14. 2021Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layerscited by 607
  15. 2020HiPPO: Recurrent Memory with Optimal Polynomial Projectionscited by 600
  16. 2017Understanding the Effective Receptive Field in Deep Convolutional Neural Networkscited by 549
  17. 2021Designing a Practical Degradation Model for Deep Blind Image Super-Resolutioncited by 524
  18. 2017FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoisingcited by 509
  19. 2020Plug-and-Play Image Restoration with Deep Denoiser Priorcited by 467
  20. 2017Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolutioncited by 452
  21. 2022Activating More Pixels in Image Super-Resolution Transformercited by 433
  22. 2024U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentationcited by 412
  23. 2017Learning Deep CNN Denoiser Prior for Image Restorationcited by 407
  24. 2015Sketch-based Manga Retrieval using Manga109 Datasetcited by 326
  25. 2024SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentationcited by 266
  26. 2020Single Image Super-Resolution via a Holistic Attention Networkcited by 265
  27. 2022Long Range Language Modeling via Gated State Spacescited by 265
  28. 2018Fast, Accurate, and Lightweight Super-Resolution with Cascading Residual Networkcited by 263
  29. 2019Lightweight Image Super-Resolution with Information Multi-distillation Networkcited by 227
  30. 2017Benchmarking Denoising Algorithms with Real Photographscited by 209
  31. 2019Feedback Network for Image Super-Resolutioncited by 155
  32. 2022Efficient Long-Range Attention Network for Image Super-resolutioncited by 151
  33. 2023Efficient and Explicit Modelling of Image Hierarchies for Image Restorationcited by 145
  34. 2015Compression Artifacts Reduction by a Deep Convolutional Networkcited by 136
  35. 2020Interpreting Super-Resolution Networks with Local Attribution Mapscited by 136
  36. 2023Selective Structured State-Spaces for Long-Form Video Understandingcited by 105
  37. 2020Cross-Scale Internal Graph Neural Network for Image Super-Resolutioncited by 94
  38. 2021LAPAR: Linearly-Assembled Pixel-Adaptive Regression Network for Single Image Super-Resolution and Beyondcited by 88

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 110; 2025: 146; 2026: 93.

349 papers.

Sort:

By area:

  1. 2024VideoMamba: State Space Model for Efficient Video Understandingcited by 325
  2. 2024PlainMamba: Improving Non-Hierarchical Mamba in Visual Recognitioncited by 141
  3. 2024RoboMamba: Efficient Vision-Language-Action Model for Robotic Reasoning and Manipulationcited by 102
  4. 2024MambaIRv2: Attentive State Space Restorationcited by 90
  5. 2024ZigMa: A DiT-style Zigzag Mamba Diffusion Modelcited by 84
  6. 2024FusionMamba: Dynamic Feature Enhancement for Multimodal Image Fusion with Mambacited by 81
  7. 2024Complexity Experts are Task-Discriminative Learners for Any Image Restorationcited by 72
  8. 2024Sigma: Siamese Mamba Network for Multi-Modal Semantic Segmentationcited by 71
  9. 2024Mamba-360: Survey of State Space Models as Transformer Alternative for Long Sequence Modelling: Methods, Applications, and Challengescited by 63
  10. 2024Multi-Scale VMamba: Hierarchy in Hierarchy Visual State Space Modelcited by 62
  11. 2025HVI: A New Color Space for Low-light Image Enhancementcited by 62
  12. 2024Perceive-IR: Learning to Perceive Degradation Better for All-in-One Image Restorationcited by 60
38 more
  1. 2024State Space Model for New-Generation Network Alternative to Transformers: A Surveycited by 59
  2. 2024A Survey on All-in-One Image Restoration: Taxonomy, Evaluation and Future Trendscited by 58
  3. 2024FreqMamba: Viewing Mamba from a Frequency Perspective for Image Derainingcited by 57
  4. 2024Vision Mamba: A Comprehensive Survey and Taxonomycited by 53
  5. 2024A Survey on Visual Mambacited by 50
  6. 2024MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Spacecited by 48
  7. 2024Frequency-Assisted Mamba for Remote Sensing Image Super-Resolutioncited by 47
  8. 2024MaIR: A Locality- and Continuity-Preserving Mamba for Image Restorationcited by 43
  9. 2024Spatial-Mamba: Effective Visual State Space Models via Structure-aware State Fusioncited by 40
  10. 2025Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networkscited by 37
  11. 2024Efficient Real-world Image Super-Resolution Via Adaptive Directional Gradient Convolutioncited by 36
  12. 2024Restore-RWKV: Efficient and Effective Medical Image Restoration with RWKVcited by 36
  13. 2024VMRNN: Integrating Vision Mamba and LSTM for Efficient and Accurate Spatiotemporal Forecastingcited by 35
  14. 2024FourierMamba: Fourier Learning Integration with State Space Models for Image Derainingcited by 32
  15. 2024VSSD: Vision Mamba with Non-Causal State Space Dualitycited by 29
  16. 2024MambaVC: Learned Visual Compression with Selective State Spacescited by 28
  17. 2024Efficient Visual State Space Model for Image Deblurringcited by 27
  18. 2024QMambaBSR: Burst Image Super-Resolution with Query State Space Modelcited by 27
  19. 2025Directing Mamba to Complex Textures: An Efficient Texture-Aware State Space Model for Image Restorationcited by 27
  20. 2024Learning Enriched Features via Selective State Spaces Model for Efficient Image Deblurringcited by 25
  21. 2025The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Reportcited by 25
  22. 2024DVMSR: Distillated Vision Mamba for Efficient Super-Resolutioncited by 23
  23. 2024VFIMamba: Video Frame Interpolation with State Space Modelscited by 22
  24. 2025Linear Attention Modeling for Learned Image Compressioncited by 22
  25. 2025From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspectivecited by 22
  26. 2024Autoregressive Pretraining with Mamba in Visioncited by 21
  27. 2025UnfoldIR: Rethinking Deep Unfolding Network in Illumination Degradation Image Restorationcited by 21
  28. 2024SSUMamba: Spatial-Spectral Selective State Space Model for Hyperspectral Image Denoisingcited by 20
  29. 2024Serp-Mamba: Advancing High-Resolution Retinal Vessel Segmentation with Selective State-Space Modelcited by 20
  30. 2025UniUIR: Considering Underwater Image Restoration as An All-in-One Learnercited by 20
  31. 2024QuadMamba: Learning Quadtree-based Selective Scan for Visual State Space Modelcited by 18
  32. 2024RGBT Tracking via All-layer Multimodal Interactions with Progressive Fusion Mambacited by 17
  33. 2024HIIF: Hierarchical Encoding based Implicit Image Function for Continuous Super-resolutioncited by 17
  34. 2024Text-controlled Motion Mamba: Text-Instructed Temporal Grounding of Human Motioncited by 16
  35. 2024Hi-Mamba: Hierarchical Mamba for Efficient Image Super-Resolutioncited by 16
  36. 2025SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structurescited by 16
  37. 2024MMR-Mamba: Multi-Modal MRI Reconstruction with Mamba and Spatial-Frequency Information Fusioncited by 15
  38. 2024GroupMamba: Efficient Group-Based Visual State Space Modelcited by 15

50 most cited of 349 All 349 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025MatIR: A Hybrid Mamba-Transformer Image Restoration Model
  2. 2024CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration
  3. 2024Multi-Scale Representation Learning for Image Restoration with State-Space Model
  4. 2024Mamba in Vision: A Comprehensive Survey of Techniques and Applications
  5. 2024VmambaIR: Visual State Space Model for Image Restoration
  6. 2024Visual Mamba: A Survey and New Outlooks
  7. 2025WaMaIR: Image Restoration via Multiscale Wavelet Convolutions and Mamba-based Channel Modeling with Texture Enhancement
  8. 2024MambaOut: Do We Really Need Mamba for Vision?