han-xudong/opencv_camera_calibration

This directory contains everything needed to deploy the Python OpenCV backend to a Hugging Face Space.

Steps to Deploy

1. 1.Create a New Space:

• —Go to Hugging Face Spaces.
• —Click "Create new Space".
• —Enter a name (e.g., camera-calibrator-backend).
• —Select Docker as the SDK.
• —Choose "Public" or "Private" (Public is easier for GitHub Pages CORS).
• —Click "Create Space".

1. 1.Upload Files:

• —You can clone the Space repository locally and copy the files from this hf_space directory into it.
• —OR, you can upload files directly via the web interface.
• —Crucial: You need to upload the cpp folder as well.
• —Structure should look like:

    /
    ├── Dockerfile
    ├── app.py
    └── requirements.txt

1. 1.Wait for Build:

• —Hugging Face will automatically build the Docker image. This might take a few minutes as it compiles OpenCV and the C++ code.
• —Check the "Logs" tab to ensure it starts successfully (you should see "Uvicorn running on ...").

1. 1.Get the API URL:

• —Once running, your API URL will be something like: https://username-space-name.hf.space.
• —You can verify it by visiting https://username-space-name.hf.space/docs to see the Swagger UI.

1. 1.Connect Frontend:
2. 2.Go to your GitHub repository settings -> Secrets and variables -> Actions.
3. 3.Add a new Repository Variable (or Secret): NEXT_PUBLIC_BACKEND_API_URL.
4. 4.Value: https://username-space-name.hf.space (no trailing slash).
5. 5.Re-run your GitHub Actions workflow to rebuild the frontend with this variable.

Troubleshooting

• —CORS Errors: If the frontend says "Network Error" or CORS issues, check app.py. Ensure your GitHub Pages domain is in the origins list.
• —Build Failures: Check the Space Build Logs. Common issues are missing dependencies (handled in Dockerfile) or C++ compilation errors.