Update README.md
Browse filesPrompt-Guided Diffusion Inpainting for License Plate Degradation is a synthetic dataset designed to simulate real-world degradations on vehicle license plates. Using a diffusion-based inpainting pipeline with ControlNet edge conditioning and prompt guidance, the dataset introduces degradations such as dust, blur, occlusion, shadows, and lighting artifacts. It includes paired clean and degraded license plate samples along with binary masks and text prompts, making it suitable for OCR benchmarking, robustness evaluation, and training more resilient Automatic Number Plate Recognition (ANPR) systems.
README.md
CHANGED
|
@@ -1,3 +1,13 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
tags:
|
| 6 |
+
- code
|
| 7 |
+
- synthetic
|
| 8 |
+
- number
|
| 9 |
+
- ANPR
|
| 10 |
+
pretty_name: ' Prompt-Guided-Diffusion-Inpainting-for-License-Plate-Degradation'
|
| 11 |
+
size_categories:
|
| 12 |
+
- 1K<n<10K
|
| 13 |
+
---
|