mlsanityData to 7 Oct 2026

Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement

arXiv 2303.06705 ↗First posted 12 Mar 2023cs.CVcited by 228

Not on the map. Why?

Cited by 21 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

When enhancing low-light images, many deep learning algorithms are based on the Retinex theory. However, the Retinex model does not consider the corruptions hidden in the dark or introduced by the light-up process. Besides, these methods usually require a tedious multi-stage training pipeline and rely on convolutional neural networks, showing limitations in capturing long-range dependencies. In this paper, we formulate a simple yet principled One-stage Retinex-based Framework (ORF). ORF first estimates the illumination information to light up the low-light image and then restores the corruption to produce the enhanced image. We design an Illumination-Guided Transformer (IGT) that utilizes illumination representations to direct the modeling of non-local interactions of regions with different lighting conditions. By plugging IGT into ORF, we obtain our algorithm, Retinexformer. Comprehensive quantitative and qualitative experiments demonstrate that our Retinexformer significantly outperforms state-of-the-art methods on thirteen benchmarks. The user study and application on low-light object detection also reveal the latent practical values of our method. Code, models, and results are available at https://github.com/caiyuanhao1998/Retinexformer

Key facts. Cited by 228 papers in the map (data through 7 Oct 2026). Builds on 35 of its 72 references that are in the map. First posted on arXiv 12 Mar 2023.

Builds on

35 of 72 references.

References · 35

  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. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  6. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  7. 2020End-to-End Object Detection with Transformerscited by 7,384
  8. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  9. 2018YOLOv3: An Incremental Improvementcited by 2,652
  10. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,869
  11. 2021Restormer: Efficient Transformer for High-Resolution Image Restorationcited by 1,618
  12. 2021ViViT: A Video Vision Transformercited by 1,610
23 more
  1. 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,368
  2. 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,218
  3. 2020Pre-Trained Image Processing Transformercited by 891
  4. 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 864
  5. 2018Deep Retinex Decomposition for Low-Light Enhancementcited by 720
  6. 2020Zero-Reference Deep Curve Estimation for Low-Light Image Enhancementcited by 541
  7. 2019EnlightenGAN: Deep Light Enhancement without Paired Supervisioncited by 466
  8. 2018Learning to See in the Darkcited by 459
  9. 2019Kindling the Darkness: A Practical Low-light Image Enhancercited by 355
  10. 2021XCiT: Cross-Covariance Image Transformerscited by 335
  11. 2020Learning Enriched Features for Real Image Restoration and Enhancementcited by 311
  12. 2020Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancementcited by 303
  13. 2015LLNet: A Deep Autoencoder Approach to Natural Low-light Image Enhancementcited by 224
  14. 2020Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-timecited by 163
  15. 2021TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Upcited by 162
  16. 2018Getting to Know Low-light Images with The Exclusively Dark Datasetcited by 157
  17. 2021StyleSwin: Transformer-based GAN for High-resolution Image Generationcited by 152
  18. 2021Generative Adversarial Transformerscited by 120
  19. 2020DeepLPF: Deep Local Parametric Filters for Image Enhancementcited by 107
  20. 2022LEDNet: Joint Low-light Enhancement and Deblurring in the Darkcited by 95
  21. 2022Unsupervised Night Image Enhancement: When Layer Decomposition Meets Light-Effects Suppressioncited by 73
  22. 2022Degradation-Aware Unfolding Half-Shuffle Transformer for Spectral Compressive Imagingcited by 66
  23. 2019Unpaired Image Enhancement Featuring Reinforcement-Learning-Controlled Image Editing Softwarecited by 34

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 8; 2024: 65; 2025: 85; 2026: 70.

228 papers.

Sort:

By area:

  1. 2024OneRestore: A Universal Restoration Framework for Composite Degradationcited by 91
  2. 2023Structure-Aware Sparse-View X-ray 3D Reconstructioncited by 70
  3. 2024LightenDiffusion: Unsupervised Low-Light Image Enhancement with Latent-Retinex Diffusion Modelscited by 70
  4. 2023Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Modelcited by 69
  5. 2025HVI: A New Color Space for Low-light Image Enhancementcited by 62
  6. 2024Perceive-IR: Learning to Perceive Degradation Better for All-in-One Image Restorationcited by 60
  7. 2024Retinexmamba: Retinex-based Mamba for Low-light Image Enhancementcited by 59
  8. 2024Zero-Reference Low-Light Enhancement via Physical Quadruple Priorscited by 58
  9. 2024Vision Mamba: A Comprehensive Survey and Taxonomycited by 53
  10. 2024Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approachcited by 48
  11. 2024MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Spacecited by 48
  12. 2024DarkIR: Robust Low-Light Image Restorationcited by 48
38 more
  1. 2024RestoreAgent: Autonomous Image Restoration Agent via Multimodal Large Language Modelscited by 41
  2. 2024NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Resultscited by 40
  3. 2024PromptFix: You Prompt and We Fix the Photocited by 40
  4. 2024HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splattingcited by 35
  5. 2024GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrievalcited by 33
  6. 2024NTIRE 2024 Challenge on Low Light Image Enhancement: Methods and Resultscited by 26
  7. 2024LYT-NET: Lightweight YUV Transformer-based Network for Low-light Image Enhancementcited by 22
  8. 2024Towards Real-world Event-guided Low-light Video Enhancement and Deblurringcited by 22
  9. 2025The Tenth NTIRE 2025 Image Denoising Challenge Reportcited by 21
  10. 2024ShadowRefiner: Towards Mask-free Shadow Removal via Fast Fourier Transformercited by 20
  11. 2024Dropout the High-rate Downsampling: A Novel Design Paradigm for UHD Image Restorationcited by 19
  12. 2025Multi-Agent Image Restorationcited by 19
  13. 2023Boosting Object Detection with Zero-Shot Day-Night Domain Adaptationcited by 17
  14. 2024AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancementcited by 16
  15. 2025URWKV: Unified RWKV Model with Multi-state Perspective for Low-light Image Restorationcited by 15
  16. 2024Multi-Object Tracking in the Darkcited by 14
  17. 2024Color Shift Estimation-and-Correction for Image Enhancementcited by 14
  18. 2025Lumina-OmniLV: A Unified Multimodal Framework for General Low-Level Visioncited by 14
  19. 2024EvLight++: Low-Light Video Enhancement with an Event Camera: A Large-Scale Real-World Dataset, Novel Method, and Morecited by 13
  20. 2024Efficient Diffusion as Low Light Enhancercited by 13
  21. 2024UniRestorer: Universal Image Restoration via Adaptively Estimating Image Degradation at Proper Granularitycited by 12
  22. 2025Low-Light Image Enhancement using Event-Based Illumination Estimationcited by 12
  23. 2025MobileIE: An Extremely Lightweight and Effective ConvNet for Real-Time Image Enhancement on Mobile Devicescited by 12
  24. 2023Learning from History: Task-agnostic Model Contrastive Learning for Image Restorationcited by 11
  25. 2024RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancementcited by 11
  26. 2025Learning to See in the Extremely Darkcited by 11
  27. 2025IRGPT: Understanding Real-world Infrared Image with Bi-cross-modal Curriculum on Large-scale Benchmarkcited by 11
  28. 2026NTIRE 2026 3D Restoration and Reconstruction in Real-world Adverse Conditions: RealX3D Challenge Resultscited by 11
  29. 2024DehazeDCT: Towards Effective Non-Homogeneous Dehazing via Deformable Convolutional Transformercited by 10
  30. 2025CWNet: Causal Wavelet Network for Low-Light Image Enhancementcited by 10
  31. 2024LoLI-Street: Benchmarking Low-Light Image Enhancement and Beyondcited by 9
  32. 2026RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Modelscited by 9
  33. 2024Wavelet-based Mamba with Fourier Adjustment for Low-light Image Enhancementcited by 8
  34. 2026CLIP-Guided Data Augmentation for Night-Time Image Dehazingcited by 8
  35. 2023Geometric-Aware Low-Light Image and Video Enhancement via Depth Guidancecited by 7
  36. 2024Domain-Adaptive 2D Human Pose Estimation via Dual Teachers in Extremely Low-Light Conditionscited by 7
  37. 2025DSwinIR: Rethinking Window-based Attention for Image Restorationcited by 7
  38. 2025EvoIR: Towards All-in-One Image Restoration via Evolutionary Frequency Modulationcited by 7

50 most cited of 228 All 228 (JSON)