Add comprehensive Chinese repository summary documentation

[Copilot]

📖 Overview

This PR adds a comprehensive Chinese summary document (仓库总结.md) for the lite.ai.toolkit repository to address the request for a repository overview in Chinese.

📝 What's Added

The new summary document provides a complete overview of the repository including:

🎯 Project Overview

• Description of lite.ai.toolkit as a lightweight C++ AI toolkit with 100+ AI models
• Coverage of computer vision tasks: object detection, face detection, face recognition, segmentation, matting, etc.

🏗️ Technical Architecture

• Multi-inference engine support: ONNXRuntime, MNN, NCNN, TNN, TensorRT
• Cross-platform compatibility: Linux, macOS, Windows, Android
• GPU acceleration: NVIDIA GPU support via TensorRT
• Modular design: Decoupled architecture for different inference engines

📊 Project Scale

• 800+ code files (417 headers + 383 source files)
• 100+ AI models across various computer vision tasks
• 500+ pretrained model files
• Comprehensive documentation (1100+ lines in README.md)

🔧 Key Features

• Unified API design across all models
• Docker deployment support with pre-built images
• Multi-threading and async inference support
• Rich examples and documentation

📚 Model Categories

1. Object Detection: YOLOv5/v6/v8, YOLOX, SSD
2. Face Detection: YOLOv5Face, UltraFace, RetinaFace
3. Face Recognition: ArcFace, CosFace, FaceNet
4. Face Landmarks: PFLD, 68 Landmarks
5. Segmentation: DeepLabV3, FCN
6. Matting: MODNet, BackgroundMatting
7. Face Restoration: GFPGAN, FaceRestoration
8. Other Applications: Style transfer, super-resolution, age/gender recognition

🚀 Usage Examples

// Object detection example
auto *yolov5 = new lite::cv::detection::YoloV5("yolov5s.onnx");
std::vector<lite::types::Boxf> detected_boxes;
yolov5->detect(image, detected_boxes);

// Face detection example
auto *ultraface = new lite::cv::face::detect::UltraFace("ultraface.onnx");
std::vector<lite::types::Boxf> face_boxes;
ultraface->detect(image, face_boxes);

🎯 Purpose

This documentation serves Chinese-speaking developers and researchers who want to:

• Understand the project's capabilities and architecture
• Get started with using the toolkit
• Contribute to the project development
• Deploy AI models in production environments

📍 File Location

The summary is added as 仓库总结.md in the root directory, making it easily accessible to users browsing the repository.

✅ Benefits

• Accessibility: Provides Chinese documentation for the large Chinese developer community
• Comprehensive: Covers all major aspects of the project in one document
• User-friendly: Well-structured with clear sections and examples
• Up-to-date: Reflects the current state of the project including recent developments

This addition enhances the project's documentation and makes it more accessible to Chinese-speaking developers worldwide.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.