vismatch-svc

相似影像辨識服務。

vismatch-svc is a Rust-based microservice for image similarity matching.

TL;DR

docker compose build
docker compose up -d

Check this guide!

Features

• Image comparison: Compare an uploaded image against a database of images to find the most similar matches.
• Image retrieval: Want more evidences? We provided a way to retrieve closest images to let you double check.
• Image upload: Add new images to a project's database.
• Docker ready: One-click™ docker compose deployment.

Getting Started

Prerequisites

• Docker needs to be installed on system.
• Since we need to build container on-site, network connection is needed.

Deployment

1. Clone the repository.
2. Build container

docker compose build

3. Start the service using Docker Compose:

   docker compose up -d

   The service will be available at http://localhost:3000.

4. The service mounts the ./image_root directory. Images are organized by project subdirectories within image_root.

   Example structure:

   image_root/
   ├── my_project/
   │   ├── image1.jpg
   │   ├── image2.png
   │   └── ...
   └── another_project/
       ├── photo.webp
       └── ...
   

API Reference

Again, Check this guide, I'm too lazy to write API documentation.

Talk is cheap, so I show you the code.

Project Structure

• src/: Rust source code.
• image_root/: Directory where images are stored (mounted volume in Docker).
• Dockerfile: Multi-stage Docker build.
• compose.yml: Docker Compose configuration.