# CharDiff-LP: A Diffusion Model with Character-Level Guidance for License Plate Image Restoration

arXiv: https://arxiv.org/abs/2510.17330 · HTML: https://mlsanity.com/paper/2510.17330/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 21 of its 43 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Placed in

- Diffusion models / [Inverse problems and image restoration](https://mlsanity.com/diffusion/inverse-problems-and-image-restoration/)

## Builds on (most cited first, 21 of 21)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks](https://mlsanity.com/paper/1809.00219/) (2018)
- [Image Super-Resolution Using Deep Convolutional Networks](https://mlsanity.com/paper/1501.00092/) (2014)
- [Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data](https://mlsanity.com/paper/2107.10833/) (2021)
- [Residual Dense Network for Image Super-Resolution](https://mlsanity.com/paper/1802.08797/) (2018)
- [SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Models](https://mlsanity.com/paper/2104.14951/) (2021)
- [Learning Texture Transformer Network for Image Super-Resolution](https://mlsanity.com/paper/2006.04139/) (2020)
- [Character Region Awareness for Text Detection](https://mlsanity.com/paper/1904.01941/) (2019)
- [Scene Text Image Super-Resolution in the Wild](https://mlsanity.com/paper/2005.03341/) (2020)
- [Text Prior Guided Scene Text Image Super-resolution](https://mlsanity.com/paper/2106.15368/) (2021)
- [Multi-Granularity Prediction for Scene Text Recognition](https://mlsanity.com/paper/2209.03592/) (2022)
- [An Efficient and Layout-Independent Automatic License Plate Recognition System Based on the YOLO detector](https://mlsanity.com/paper/1909.01754/) (2019)
- [A Text Attention Network for Spatial Deformation Robust Scene Text Image Super-resolution](https://mlsanity.com/paper/2203.09388/) (2022)
- [Scene Text Image Super-resolution based on Text-conditional Diffusion Models](https://mlsanity.com/paper/2311.09759/) (2023)
- [Super-Resolution of License Plate Images Using Attention Modules and Sub-Pixel Convolution Layers](https://mlsanity.com/paper/2305.17313/) (2023)
- [Forensic License Plate Recognition with Compression-Informed Transformers](https://mlsanity.com/paper/2207.14686/) (2022)
- [Enhancing License Plate Super-Resolution: A Layout-Aware and Character-Driven Approach](https://mlsanity.com/paper/2408.15103/) (2024)
- [Recognition-Guided Diffusion Model for Scene Text Image Super-Resolution](https://mlsanity.com/paper/2311.13317/) (2023)
- [MF-LPR2: Multi-Frame License Plate Image Restoration and Recognition using Optical Flow](https://mlsanity.com/paper/2508.14797/) (2025)

## Abstract (arXiv)

License plate image restoration is important not only as a preprocessing step for license plate recognition but also for enhancing evidential value, improving visual clarity, and enabling broader reuse of license plate images. We propose a novel diffusion-based framework with character-level guidance, CharDiff-LP, which effectively restores and recognizes severely degraded license plate images captured under realistic conditions. CharDiff-LP leverages fine-grained character-level priors extracted through external segmentation and Optical Character Recognition (OCR) modules tailored for low-quality license plate images. For precise and focused guidance, CharDiff-LP incorporates a novel Character-guided Attention through Region-wise Masking (CHARM) module, which ensures that each character's guidance is restricted to its own region, thereby avoiding interference with other regions. In experiments, CharDiff-LP significantly outperformed baseline restoration models in both restoration quality and recognition accuracy, achieving a 28.3% relative reduction in character error rate (CER) on the Roboflow-LP dataset compared with the best-performing baseline.