Armenian OCR
Collection
A set of models and dataset for Armenian OCR. • 2 items • Updated • 1
paddle-calfa, open Armenian OCR models (PaddleOCR)
paddle-calfa-tiny and paddle-calfa-small are open text-recognition models
for Classical, Western and Eastern Armenian, trained by
Calfa. They are PP-OCRv6 recognition models in
PaddleOCR inference format (inference.json, inference.pdiparams,
inference.yml). Each model carries its own character dictionary.
| Model | Based on | Directory | Since |
|---|---|---|---|
paddle-calfa-tiny |
PP-OCRv6-tiny | tiny/inference/ |
v1.0.0 |
paddle-calfa-small |
PP-OCRv6-small (more accurate, slower) | small/inference/ |
v1.1.0 |
- Version: v1.1.0
- License: CC BY-NC 4.0
Use with the open pipeline
These models are used by the open Armenian OCR pipeline (code, CLI and web app): https://github.com/calfa-co/hye-open-ocr
Use directly with PaddleOCR
Each inference/ directory is a standard PaddleOCR text-recognition inference
model. To run it directly with PaddleOCR (as a custom rec model), follow the
official documentation:
PaddleOCR.
- Downloads last month
- -