This Gradio demo provides an interactive interface for the Unlimited OCR model from Baidu. It is an improved version of the original Hugging Face Space demo at https://huggingface.co/spaces/baidu/Unlimited-OCR, with a more polished experience for streaming OCR output, live preview annotations, and PDF processing.
- Official model repository: https://github.com/baidu/Unlimited-OCR
- Original Hugging Face Space: https://huggingface.co/spaces/baidu/Unlimited-OCR
- Official Hugging Face model page: https://huggingface.co/baidu/Unlimited-OCR
- This demo repository for this version of the app is: https://github.com/bangonkali/unlimited-ocr-demo
- This project is built as a demo-focused improvement on the original Hugging Face Space experience.
- Runs the Baidu Unlimited OCR model via Hugging Face Transformers
- Supports single-image and PDF-based document parsing
- Streams OCR text progressively while inference is running
- Shows detected text regions with preview overlays on the source image
The implementation in this demo follows the official model usage patterns and dependency versions described by the official Unlimited OCR model page, while adapting them into a Gradio-based interactive experience.