pdfmarkdownapp's picture
Upload folder using huggingface_hub
1c66a97 verified
|
Raw
History Blame Contribute Delete
1.46 kB
---
license: apache-2.0
tags:
- ocr
- paddleocr
- pp-ocrv5
- onnx
---
# ppu-paddle-ocr-models (PP-OCRv5 mobile, multilingual)
PP-OCRv5 **mobile** (multilingual) ONNX model weights + character dictionary, mirrored as a **fallback CDN** for [pdfmarkdown.app](https://pdfmarkdown.app)'s in-browser OCR. The primary CDN is jsDelivr (npm package `pdfmarkdown-ppocrv5-models`); HuggingFace's `resolve/main` serves as the second tier when jsDelivr is unavailable.
## Files
| Path | Purpose |
| --- | --- |
| `detection/PP-OCRv5_mobile_det_infer.ort` | Text **detection** model (ORT format) |
| `recognition/PP-OCRv5_mobile_rec_infer.onnx` | Text **recognition** model (ONNX, multilingual) |
| `recognition/ppocrv5_dict.txt` | Recognition **character dictionary** |
## Resolve URLs
```
https://huggingface.co/<user>/ppu-paddle-ocr-models/resolve/main/detection/PP-OCRv5_mobile_det_infer.ort
https://huggingface.co/<user>/ppu-paddle-ocr-models/resolve/main/recognition/PP-OCRv5_mobile_rec_infer.onnx
https://huggingface.co/<user>/ppu-paddle-ocr-models/resolve/main/recognition/ppocrv5_dict.txt
```
## Attribution & License
Models are **PP-OCRv5** from [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR), **Apache-2.0**. ONNX/ORT exports and dictionary sourced from [PT-Perkasa-Pilar-Utama/ppu-paddle-ocr-models](https://github.com/PT-Perkasa-Pilar-Utama/ppu-paddle-ocr-models) (Apache-2.0), redistributed unmodified for pdfmarkdown.app's browser OCR.