hiroi-sora/PaddleOCR-json

Health v2 · maintenance only

Adoption not part of the score

1542 stars · 203 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

An offline OCR command-line executable compiled from PaddleOCR C++ that recognizes text in images and outputs results as JSON strings. It can be invoked as a subprocess or over TCP, with ready-made APIs for Python and other languages.

Use cases

• extract text from images offline
• add OCR to my app without writing C++
• batch convert screenshots to text
• recognize text from clipboard images
• call OCR over TCP from another program
• read text from handwriting or stylized fonts

When to choose

• you need fast, offline OCR with no Python runtime or internet connection
• you want a drop-in OCR engine callable from any language via subprocess or TCP
• you need good accuracy on hard cases like handwriting, small text, or cluttered backgrounds
• you have an AVX-capable x86-64 CPU on Windows or Linux

When to avoid

• your CPU lacks the AVX instruction set (use RapidOCR- instead)
• you need GPU acceleration or a lightweight deployment footprint (369MB unpacked)
• you need a pure Python OCR pipeline you can modify directly

Facets

cli-tool · maturity active

ocr cli image-processing computer-vision image-processing developer-tools windows cli paddleocr offline-ocr text-recognition subprocess -output tcp-server python-api command-line linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="hiroi-sora/PaddleOCR-json")