01 / OVERVIEW
Overview
PaddleSharp brings Paddle Inference and related model workflows into .NET. Its repository documents OCR, detection, orientation, Chinese segmentation, and model conversion. Choose packages, models, and runtimes for the specific task.
02 / CAPABILITIES
Key capabilities
01
.NET bindings for the Paddle Inference C API
02
OCR, table recognition, and supported detection models
03
Platform-specific CPU / GPU runtime packages
03 / POSSIBILITIES
Ideas to explore
- Integrate OCR and table recognition into .NET document workflows
- Evaluate Paddle models across CPU / GPU environments
- Explore Chinese text and image-model integration
04 / GET STARTED
Getting started
Before you begin
- A compatible .NET environment, native runtime, and models
- Matching CUDA and drivers for GPU packages
Get the project
dotnet add package Sdcb.PaddleOCRNext steps
- Choose task-specific core, model, and runtime packages using the documentation.
- Start with a minimal example and validate your own images and environment.
05 / SCOPE & STATUS
Current boundaries
This wraps native Paddle Inference and requires matching platform, architecture, and runtime packages. It is not a native-dependency-free C# inference engine.
This introduction is based on the public README. Consult the repository for current versions, licenses, and supported features.
Updated 2026-10-27