mlsanityData to 7 Oct 2026

Document Rectification and Illumination Correction using a Patch-based CNN

arXiv 1909.09470 ↗First posted 20 Sep 2019cs.CVcited by 37

Not on the map. Why?

Abstract (arXiv)

We propose a novel learning method to rectify document images with various distortion types from a single input image. As opposed to previous learning-based methods, our approach seeks to first learn the distortion flow on input image patches rather than the entire image. We then present a robust technique to stitch the patch results into the rectified document by processing in the gradient domain. Furthermore, we propose a second network to correct the uneven illumination, further improving the readability and OCR accuracy. Due to the less complex distortion present on the smaller image patches, our patch-based approach followed by stitching and illumination correction can significantly improve the overall accuracy in both the synthetic and real datasets.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 9; 2023: 7; 2024: 2; 2025: 11; 2026: 2.

37 papers.

Sort:

By area:

  1. 2021DocTr: Document Image Transformer for Geometric Unwarping and Illumination Correctioncited by 38
  2. 2022Geometric Representation Learning for Document Image Rectificationcited by 26
  3. 2021DocScanner: Robust Document Image Rectification with Progressive Learningcited by 24
  4. 2022Revisiting Document Image Dewarping by Grid Regularizationcited by 24
  5. 2022Fourier Document Restoration for Robust Document Dewarping and Recognitioncited by 21
  6. 2021Dewarping Document Image By Displacement Flow Estimation with Fully Convolutional Networkcited by 20
  7. 2022Document Dewarping with Control Pointscited by 19
  8. 2023UVDoc: Neural Grid-based Document Unwarpingcited by 16
  9. 2022Marior: Margin Removal and Iterative Content Rectification for Document Dewarping in the Wildcited by 15
  10. 2022UDoc-GAN: Unpaired Document Illumination Correction with Background Light Priorcited by 15
  11. 2023Deep Unrestricted Document Image Rectificationcited by 15
  12. 2023SimFIR: A Simple Framework for Fisheye Image Rectification with Self-supervised Representation Learningcited by 13
25 more
  1. 2024High-Fidelity Document Stain Removal via A Large-Scale Real-World Dataset and A Memory-Augmented Transformercited by 12
  2. 2024Visual Text Meets Low-level Vision: A Comprehensive Survey on Visual Text Processingcited by 8
  3. 2021A Survey on Deep learning based Document Image Enhancementcited by 7
  4. 2021Light-weight Document Image Cleanup using Perceptual Losscited by 6
  5. 2022Water Simulation and Rendering from a Still Photographcited by 5
  6. 2025Limb-Aware Virtual Try-On Network with Progressive Clothing Warpingcited by 5
  7. 2021Correcting Face Distortion in Wide-Angle Videoscited by 4
  8. 2022Geometric Rectification of Creased Document Images based on Isometric Mappingcited by 4
  9. 2023DocMAE: Document Image Rectification via Self-supervised Representation Learningcited by 4
  10. 2025DvD: Unleashing a Generative Paradigm for Document Dewarping via Coordinates-based Diffusion Modelcited by 4
  11. 2025Uni-DocDiff: A Unified Document Restoration Model Based on Diffusioncited by 4
  12. 2023DocAligner: Annotating Real-world Photographic Document Images by Simply Taking Picturescited by 3
  13. 2023MataDoc: Margin and Text Aware Document Dewarping for Arbitrary Boundarycited by 3
  14. 2025Progressive Limb-Aware Virtual Try-Oncited by 3
  15. 2025Toward Real-world Text Image Forgery Localization: Structured and Interpretable Data Synthesiscited by 3
  16. 2022Digitizing Historical Balance Sheet Data: A Practitioner's Guidecited by 1
  17. 2025ForCenNet: Foreground-Centric Network for Document Image Rectificationcited by 1
  18. 2023Unfolder: Fast localization and image rectification of a document with a crease from folding in halfcited by 0
  19. 2025Document Image Rectification Bases on Self-Adaptive Multitask Fusioncited by 0
  20. 2025GL-PGENet: A Parameterized Generation Framework for Robust Document Image Enhancementcited by 0
  21. 2025DocShaDiffusion: Diffusion Model in Latent Space for Document Image Shadow Removalcited by 0
  22. 2025DocIQ: A Benchmark Dataset and Feature Fusion Network for Document Image Quality Assessmentcited by 0
  23. 2025Cascaded Robust Rectification for Arbitrary Document Imagescited by 0
  24. 2026SyntheticDoc: A Large Synthetic Dataset for Document Unwarping and Illumination Correctioncited by 0
  25. 2026ENet-GP: Unified Document Image Restorationcited by 0