OnnxOCR
Lightweight OCR model with fast inference speed
- Supports OCR recognition in over 80 languages
- Convert to ONNX model to improve inference speed
- Independent of deep learning training frameworks, easy to deploy
- Suitable for ARM and x86 architecture computers
- On computers with equivalent performance, the inference speed is increased by 4-5 times
- Suitable for scenarios where computing power is limited but accuracy needs to be maintained
Product Details
OnnxOCR is a lightweight OCR model reconstructed based on PaddleOCR, which breaks away from the PaddlePaddle deep learning training framework and achieves fast inference speed. This model supports inference in over 80 languages and, when converted to ONNX models, its inference speed is 5 times faster than using PaddlePaddle framework. OnnxOCR is independent of deep learning training frameworks and can be deployed directly. It is suitable for scenarios with limited computing power but requires accuracy, and can be deployed on ARM and x86 architecture computers.
